
# Import required libraries
install.packages("GGally")
install.packages("matrixStats")
install.packages("readxl")
install.packages("lubridate")
install.packages("writexl")
library(readxl)
library(forecast)
library(data.table)
library(lubridate)
library(dplyr)
library(tidyr)
library(GGally)
library(dplyr)
library(writexl)
library(matrixStats)

The downloaded binary packages are in
	/var/folders/tm/h12dzwn519q9gwvz38d0m4_80000gn/T//RtmpqbESn6/downloaded_packages

The downloaded binary packages are in
	/var/folders/tm/h12dzwn519q9gwvz38d0m4_80000gn/T//RtmpqbESn6/downloaded_packages

The downloaded binary packages are in
	/var/folders/tm/h12dzwn519q9gwvz38d0m4_80000gn/T//RtmpqbESn6/downloaded_packages

The downloaded binary packages are in
	/var/folders/tm/h12dzwn519q9gwvz38d0m4_80000gn/T//RtmpqbESn6/downloaded_packages

The downloaded binary packages are in
	/var/folders/tm/h12dzwn519q9gwvz38d0m4_80000gn/T//RtmpqbESn6/downloaded_packages


tday=today("Turkey")
day_before_tday <- tday - 1
#day_before_tday <- tday - 2
prediction_day <- tday +1
start_date <- as.Date("2024-02-01")
end_date <- as.Date("2024-05-15")


file_weather = paste0("/Users/ecemozturk/Desktop/processed_weather.csv")
file_production = paste0("/Users/ecemozturk/Desktop/production-4.csv")
weather_data = fread(file_weather)
production_data = fread(file_production)


# getting full weather date and hours as a template
template_dt = unique(weather_data[,list(date,hour)])
template_dt = merge(template_dt,production_data,by=c('date','hour'),all.x=T)
#template_dt = template_dt[date<=(tday + 1)]
template_dt = template_dt[date<=(tday)]


###NA VALUES###
any_na <- anyNA(weather_data)
if (any_na) {
  cat("The dataset contains NA values.\n")
  # Display the count of NAs per column
  print(colSums(is.na(weather_data)))
} else {
  cat("The dataset does not contain any NA values.\n")
}
# Display all rows that have NA values
na_rows <- weather_data[!complete.cases(weather_data), ]
View(na_rows)

# Fill NA values with the average of the surrounding values (linear interpolation)
merged_data_filled <- weather_data %>%
  mutate(across(where(is.numeric), ~ na.approx(.x, na.rm = FALSE)))

# Fill leading NAs with the next available value, upward
merged_data_filled <- merged_data_filled %>%
  mutate(across(where(is.numeric), ~ na.locf(.x, fromLast = TRUE, na.rm = FALSE)))

weather_data<- merged_data_filled

The dataset contains NA values.
                   date                    hour                     lat 
                      0                       0                       0 
                    lon           dswrf_surface    tcdc_low.cloud.layer 
                      0                      25                      50 
tcdc_middle.cloud.layer   tcdc_high.cloud.layer  tcdc_entire.atmosphere 
                     50                     100                      63 
uswrf_top_of_atmosphere           csnow_surface           dlwrf_surface 
                     75                      25                      25 
          uswrf_surface             tmp_surface 
                     50                      25

Error in `mutate()`:
ℹ In argument: `across(where(is.numeric), ~na.approx(.x, na.rm =
  FALSE))`.
Caused by error in `across()`:
! Can't compute column `hour`.
Caused by error in `na.approx()`:
! "na.approx" fonksiyonu bulunamadı
Traceback:

1. weather_data %>% mutate(across(where(is.numeric), ~na.approx(.x, 
 .     na.rm = FALSE)))
2. mutate(., across(where(is.numeric), ~na.approx(.x, na.rm = FALSE)))
3. mutate.data.frame(., across(where(is.numeric), ~na.approx(.x, 
 .     na.rm = FALSE)))
4. mutate_cols(.data, dplyr_quosures(...), by)
5. withCallingHandlers(for (i in seq_along(dots)) {
 .     poke_error_context(dots, i, mask = mask)
 .     context_poke("column", old_current_column)
 .     new_columns <- mutate_col(dots[[i]], data, mask, new_columns)
 . }, error = dplyr_error_handler(dots = dots, mask = mask, bullets = mutate_bullets, 
 .     error_call = error_call, error_class = "dplyr:::mutate_error"), 
 .     warning = dplyr_warning_handler(state = warnings_state, mask = mask, 
 .         error_call = error_call))
6. mutate_col(dots[[i]], data, mask, new_columns)
7. withCallingHandlers(mask$eval_all_mutate(quo), error = function(cnd) {
 .     name <- dplyr_quosure_name(quo_data)
 .     msg <- glue("Can't compute column `{name}`.")
 .     abort(msg, call = call("across"), parent = cnd)
 . })
8. mask$eval_all_mutate(quo)
9. eval()
10. .handleSimpleError(function (cnd) 
  . {
  .     name <- dplyr_quosure_name(quo_data)
  .     msg <- glue("Can't compute column `{name}`.")
  .     abort(msg, call = call("across"), parent = cnd)
  . }, "\"na.approx\" fonksiyonu bulunamadı", base::quote(na.approx(hour, 
  .     na.rm = FALSE)))
11. h(simpleError(msg, call))
12. abort(msg, call = call("across"), parent = cnd)
13. signal_abort(cnd, .file)
14. signalCondition(cnd)
15. (function (cnd) 
  . {
  .     local_error_context(dots, i = frame[[i_sym]], mask = mask)
  .     if (inherits(cnd, "dplyr:::internal_error")) {
  .         parent <- error_cnd(message = bullets(cnd))
  .     }
  .     else {
  .         parent <- cnd
  .     }
  .     message <- c(cnd_bullet_header(action), i = if (has_active_group_context(mask)) cnd_bullet_cur_group_label())
  .     abort(message, class = error_class, parent = parent, call = error_call)
  . })(structure(list(message = structure("Can't compute column `hour`.", class = c("glue", 
  . "character")), trace = structure(list(call = list(IRkernel::main(), 
  .     kernel$run(), handle_shell(), executor$execute(msg), tryCatch(evaluate(request$content$code, 
  .         envir = .GlobalEnv, output_handler = oh, stop_on_error = 1L), 
  .         interrupt = function(cond) {
  .             log_debug("Interrupt during execution")
  .             interrupted <<- TRUE
  .         }, error = .self$handle_error), tryCatchList(expr, classes, 
  .         parentenv, handlers), tryCatchOne(tryCatchList(expr, 
  .         names[-nh], parentenv, handlers[-nh]), names[nh], parentenv, 
  .         handlers[[nh]]), doTryCatch(return(expr), name, parentenv, 
  .         handler), tryCatchList(expr, names[-nh], parentenv, handlers[-nh]), 
  .     tryCatchOne(expr, names, parentenv, handlers[[1L]]), doTryCatch(return(expr), 
  .         name, parentenv, handler), evaluate(request$content$code, 
  .         envir = .GlobalEnv, output_handler = oh, stop_on_error = 1L), 
  .     evaluate_call(expr, parsed$src[[i]], envir = envir, enclos = enclos, 
  .         debug = debug, last = i == length(out), use_try = stop_on_error != 
  .             2L, keep_warning = keep_warning, keep_message = keep_message, 
  .         log_echo = log_echo, log_warning = log_warning, output_handler = output_handler, 
  .         include_timing = include_timing), timing_fn(handle(ev <- withCallingHandlers(withVisible(eval_with_user_handlers(expr, 
  .         envir, enclos, user_handlers)), warning = wHandler, error = eHandler, 
  .         message = mHandler))), handle(ev <- withCallingHandlers(withVisible(eval_with_user_handlers(expr, 
  .         envir, enclos, user_handlers)), warning = wHandler, error = eHandler, 
  .         message = mHandler)), try(f, silent = TRUE), tryCatch(expr, 
  .         error = function(e) {
  .             call <- conditionCall(e)
  .             if (!is.null(call)) {
  .                 if (identical(call[[1L]], quote(doTryCatch))) 
  .                   call <- sys.call(-4L)
  .                 dcall <- deparse(call, nlines = 1L)
  .                 prefix <- paste("Error in", dcall, ": ")
  .                 LONG <- 75L
  .                 sm <- strsplit(conditionMessage(e), "\n")[[1L]]
  .                 w <- 14L + nchar(dcall, type = "w") + nchar(sm[1L], 
  .                   type = "w")
  .                 if (is.na(w)) 
  .                   w <- 14L + nchar(dcall, type = "b") + nchar(sm[1L], 
  .                     type = "b")
  .                 if (w > LONG) 
  .                   prefix <- paste0(prefix, "\n  ")
  .             }
  .             else prefix <- "Error : "
  .             msg <- paste0(prefix, conditionMessage(e), "\n")
  .             .Internal(seterrmessage(msg[1L]))
  .             if (!silent && isTRUE(getOption("show.error.messages"))) {
  .                 cat(msg, file = outFile)
  .                 .Internal(printDeferredWarnings())
  .             }
  .             invisible(structure(msg, class = "try-error", condition = e))
  .         }), tryCatchList(expr, classes, parentenv, handlers), 
  .     tryCatchOne(expr, names, parentenv, handlers[[1L]]), doTryCatch(return(expr), 
  .         name, parentenv, handler), withCallingHandlers(withVisible(eval_with_user_handlers(expr, 
  .         envir, enclos, user_handlers)), warning = wHandler, error = eHandler, 
  .         message = mHandler), withVisible(eval_with_user_handlers(expr, 
  .         envir, enclos, user_handlers)), eval_with_user_handlers(expr, 
  .         envir, enclos, user_handlers), eval(expr, envir, enclos), 
  .     eval(expr, envir, enclos), weather_data %>% mutate(across(where(is.numeric), 
  .         ~na.approx(.x, na.rm = FALSE))), mutate(., across(where(is.numeric), 
  .         ~na.approx(.x, na.rm = FALSE))), mutate.data.frame(., 
  .         across(where(is.numeric), ~na.approx(.x, na.rm = FALSE))), 
  .     mutate_cols(.data, dplyr_quosures(...), by), withCallingHandlers(for (i in seq_along(dots)) {
  .         poke_error_context(dots, i, mask = mask)
  .         context_poke("column", old_current_column)
  .         new_columns <- mutate_col(dots[[i]], data, mask, new_columns)
  .     }, error = dplyr_error_handler(dots = dots, mask = mask, 
  .         bullets = mutate_bullets, error_call = error_call, error_class = "dplyr:::mutate_error"), 
  .         warning = dplyr_warning_handler(state = warnings_state, 
  .             mask = mask, error_call = error_call)), mutate_col(dots[[i]], 
  .         data, mask, new_columns), withCallingHandlers(mask$eval_all_mutate(quo), 
  .         error = function(cnd) {
  .             name <- dplyr_quosure_name(quo_data)
  .             msg <- glue("Can't compute column `{name}`.")
  .             abort(msg, call = call("across"), parent = cnd)
  .         }), mask$eval_all_mutate(quo), eval(), .handleSimpleError(`<fn>`, 
  .         "\"na.approx\" fonksiyonu bulunamadı", base::quote(na.approx(hour, 
  .             na.rm = FALSE))), h(simpleError(msg, call)), abort(msg, 
  .         call = call("across"), parent = cnd)), parent = c(0L, 
  . 1L, 2L, 3L, 4L, 5L, 6L, 7L, 6L, 9L, 10L, 4L, 12L, 13L, 13L, 15L, 
  . 16L, 17L, 18L, 19L, 13L, 13L, 13L, 23L, 24L, 0L, 0L, 0L, 28L, 
  . 29L, 29L, 31L, 31L, 33L, 0L, 35L, 36L), visible = c(TRUE, TRUE, 
  . TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, 
  . TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, 
  . TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, FALSE, 
  . FALSE, FALSE), namespace = c("IRkernel", NA, "IRkernel", NA, 
  . "base", "base", "base", "base", "base", "base", "base", "evaluate", 
  . "evaluate", "evaluate", "evaluate", "base", "base", "base", "base", 
  . "base", "base", "base", "evaluate", "base", "base", NA, "dplyr", 
  . "dplyr", "dplyr", "base", "dplyr", "base", NA, "dplyr", "base", 
  . "dplyr", "rlang"), scope = c("::", NA, "local", NA, "::", "local", 
  . "local", "local", "local", "local", "local", "::", ":::", "local", 
  . "local", "::", "::", "local", "local", "local", "::", "::", ":::", 
  . "::", "::", NA, "::", ":::", ":::", "::", ":::", "::", NA, "local", 
  . "::", "local", "::")), row.names = c(NA, -37L), version = 2L, class = c("rlang_trace", 
  . "rlib_trace", "tbl", "data.frame")), parent = structure(list(
  .     message = "\"na.approx\" fonksiyonu bulunamadı", call = na.approx(hour, 
  .         na.rm = FALSE)), class = c("simpleError", "error", "condition"
  . )), rlang = list(inherit = TRUE), call = across(), use_cli_format = TRUE), class = c("rlang_error", 
  . "error", "condition")))
16. abort(message, class = error_class, parent = parent, call = error_call)
17. signal_abort(cnd, .file)


#Coordinate aggregation by long to wide format
long_weather <- weather_data
long_weather <- melt(weather_data,id.vars=c(1:4))
hourly_region_averages = dcast(long_weather, date+hour~variable,fun.aggregate=mean)
View(hourly_region_averages)

Warning message in melt.data.table(weather_data, id.vars = c(1:4)):
“'measure.vars' [dswrf_surface, tcdc_low.cloud.layer, tcdc_middle.cloud.layer, tcdc_high.cloud.layer, ...] are not all of the same type. By order of hierarchy, the molten data value column will be of type 'double'. All measure variables not of type 'double' will be coerced too. Check DETAILS in ?melt.data.table for more on coercion.”


# Merge with hourly_region_averages
template_dt_with_weather <- merge(template_dt, hourly_region_averages, by = c('date', 'hour'), all.x = TRUE)
View(template_dt_with_weather)
#Order it by date and hour
template_dt_with_weather = template_dt_with_weather[order(date,hour)]

template_dt_with_aggregate <- template_dt_with_weather

template_dt_with_aggregate$hourly_cloud_average <- rowMeans(select(template_dt_with_aggregate, starts_with("tcdc_")), na.rm = TRUE)
template_dt_with_aggregate$hourly_max_t <- rowMaxs(as.matrix(select(template_dt_with_aggregate, starts_with("tmp_"))), na.rm = TRUE)
View(template_dt_with_aggregate)


# Use select to exclude coumns starting with "tcdc_" to focus on average
template_dt_with_aggregate <- template_dt_with_aggregate %>%
  select(-starts_with("tcdc_"))
#-starts_with("tmp_"))

# Read your holiday dataset
#holiday_data <- read.csv("/Users/ecemozturk/Desktop/Short Holiday.csv")
holiday_data <- read_excel("/Users/ecemozturk/Desktop/Short Holiday12.xlsx")
# Convert date column to Date format
holiday_data$date <- as.Date(holiday_data$date, format = "%d.%m.%Y", na.rm = TRUE)

# Merge holiday data with production dataset based on the date column
all_data <- merge(template_dt_with_aggregate, holiday_data, by = "date", all.x = TRUE)


all_data = all_data[!is.na(production)]
all_data_daily <- all_data[all_data$date == day_before_tday, ]

available_data = all_data[!is.na(production) & hour >= 4 & hour <= 19,]
#to_be_forecasted = template_dt_with_weather[is.na(production)]
to_be_forecasted <- all_data[is.na(production) & hour >= 4 & hour <= 19, ]

View(all_data)


# Plot production data as a line graph
plot(all_data$date, all_data$production, type = "l", xlab = "Date", ylab = "Production", main = "Production Data")


# Convert wday and month to characters
all_data[, wday := as.character(wday(date, label = TRUE))]
all_data[, month := as.character(month(date, label = TRUE))]

#special period was defined to represent the period when the capacity reaches up to 12.
all_data$special_period <- as.numeric(all_data$date >= as.Date("2022-06-01") & all_data$date <= as.Date("2022-11-16"))

#Linear regression for cummulative production
lm_model <- lm(production ~.+hourly_max_t*month+wday  -date, data = all_data)
summary(lm_model)

Call:
lm(formula = production ~ . + hourly_max_t * month + wday - date, 
    data = all_data)

Residuals:
    Min      1Q  Median      3Q     Max 
-9.2566 -1.3381 -0.2832  1.1297 10.1386 

Coefficients: (2 not defined because of singularities)
                          Estimate Std. Error t value Pr(>|t|)    
(Intercept)             -1.183e+02  1.806e+00 -65.491  < 2e-16 ***
hour                    -7.434e-02  2.426e-03 -30.643  < 2e-16 ***
dswrf_surface           -8.027e-03  2.250e-04 -35.682  < 2e-16 ***
uswrf_top_of_atmosphere  6.639e-03  2.504e-04  26.511  < 2e-16 ***
csnow_surface            2.107e+00  1.078e-01  19.540  < 2e-16 ***
dlwrf_surface           -8.416e-02  9.777e-04 -86.087  < 2e-16 ***
uswrf_surface            1.251e-02  6.585e-04  19.005  < 2e-16 ***
tmp_surface              5.074e-01  6.415e-03  79.098  < 2e-16 ***
hourly_cloud_average     2.601e-02  9.883e-04  26.316  < 2e-16 ***
hourly_max_t                    NA         NA      NA       NA    
is.weekend               4.192e-01  2.437e-01   1.720 0.085457 .  
is.ramadan              -3.223e-01  6.984e-02  -4.615 3.96e-06 ***
is.religousday           1.978e-01  9.355e-02   2.114 0.034518 *  
is.nationalday           2.637e-01  1.608e-01   1.640 0.101032    
is.publicholiday        -6.239e-01  2.425e-01  -2.573 0.010102 *  
wdayCum                 -1.783e-01  5.583e-02  -3.193 0.001409 ** 
wdayÇar                 -1.642e-01  5.564e-02  -2.950 0.003180 ** 
wdayPaz                 -1.141e-01  5.577e-02  -2.046 0.040762 *  
wdayPer                 -2.192e-01  5.578e-02  -3.929 8.54e-05 ***
wdayPts                 -1.291e-01  5.565e-02  -2.320 0.020357 *  
wdaySal                         NA         NA      NA       NA    
monthAra                -7.767e+00  3.301e+00  -2.353 0.018632 *  
monthEki                 1.333e-01  2.530e+00   0.053 0.957989    
monthEyl                -4.919e-01  2.140e+00  -0.230 0.818168    
monthHaz                -3.846e+01  2.665e+00 -14.432  < 2e-16 ***
monthKas                 4.319e+00  2.868e+00   1.506 0.132159    
monthMar                 9.223e+00  2.223e+00   4.149 3.36e-05 ***
monthMay                -2.338e+01  2.367e+00  -9.876  < 2e-16 ***
monthNis                -8.559e+00  2.200e+00  -3.891 0.000100 ***
monthOca                 7.182e+00  2.500e+00   2.873 0.004070 ** 
monthŞub                 2.010e+01  2.354e+00   8.537  < 2e-16 ***
monthTem                -1.940e+01  2.222e+00  -8.731  < 2e-16 ***
special_period           2.101e-01  4.691e-02   4.479 7.54e-06 ***
hourly_max_t:monthAra    3.382e-02  1.174e-02   2.881 0.003962 ** 
hourly_max_t:monthEki    5.904e-04  8.679e-03   0.068 0.945771    
hourly_max_t:monthEyl    1.397e-03  7.209e-03   0.194 0.846370    
hourly_max_t:monthHaz    1.356e-01  9.053e-03  14.979  < 2e-16 ***
hourly_max_t:monthKas   -1.199e-02  1.002e-02  -1.197 0.231334    
hourly_max_t:monthMar   -2.815e-02  7.718e-03  -3.647 0.000266 ***
hourly_max_t:monthMay    8.451e-02  8.114e-03  10.415  < 2e-16 ***
hourly_max_t:monthNis    3.230e-02  7.521e-03   4.294 1.76e-05 ***
hourly_max_t:monthOca   -2.041e-02  8.828e-03  -2.312 0.020798 *  
hourly_max_t:monthŞub   -6.574e-02  8.272e-03  -7.948 1.99e-15 ***
hourly_max_t:monthTem    6.668e-02  7.466e-03   8.931  < 2e-16 ***
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 2.144 on 20735 degrees of freedom
  (3 observations deleted due to missingness)
Multiple R-squared:  0.664,	Adjusted R-squared:  0.6633 
F-statistic: 999.3 on 41 and 20735 DF,  p-value: < 2.2e-16


available_data <- all_data[!is.na(production) & hour >= 4 & hour <= 18 & date < start_date,]
to_be_forecasted <- all_data[!is.na(production) & hour >= 4 & hour <= 18 & date >= start_date & date <= end_date, ]

library(dplyr)


# Filter data for hour 4
hour_4_data <- all_data[all_data$hour == 4, ]
hour_4_data$trend_hour_4 <- 1:nrow(hour_4_data)
hour_4_data[, lag_1_production := shift(production,1)]
hour_4_data[,lag_1_diff:=production-lag_1_production]
hour_4_data <- hour_4_data[!is.na(lag_1_production)]
# Fit linear regression model for hour 4
lm_hour_4 <- lm(production ~+lag_1_production +dlwrf_surface+tmp_surface+hourly_cloud_average+special_period+trend_hour_4+month, data = hour_4_data)
# Summarize the model
summary(lm_hour_4)
checkresiduals(lm_hour_4)

Call:
lm(formula = production ~ +lag_1_production + dlwrf_surface + 
    tmp_surface + hourly_cloud_average + special_period + trend_hour_4 + 
    month, data = hour_4_data)

Residuals:
      Min        1Q    Median        3Q       Max 
-0.121351 -0.001939 -0.000402  0.000778  0.173824 

Coefficients:
                       Estimate Std. Error t value Pr(>|t|)    
(Intercept)           5.239e-02  5.513e-02   0.950  0.34219    
lag_1_production      7.021e-01  2.467e-02  28.459  < 2e-16 ***
dlwrf_surface         1.859e-05  3.902e-05   0.476  0.63393    
tmp_surface          -1.980e-04  2.282e-04  -0.868  0.38588    
hourly_cloud_average -1.778e-05  2.921e-05  -0.609  0.54286    
special_period       -5.201e-03  1.516e-03  -3.431  0.00063 ***
trend_hour_4          3.035e-06  2.114e-06   1.436  0.15137    
monthAra             -4.042e-03  3.027e-03  -1.335  0.18217    
monthEki             -5.109e-04  2.435e-03  -0.210  0.83385    
monthEyl             -2.670e-04  2.246e-03  -0.119  0.90539    
monthHaz              1.294e-02  2.430e-03   5.325 1.29e-07 ***
monthKas             -2.199e-03  2.750e-03  -0.800  0.42413    
monthMar             -3.781e-03  3.050e-03  -1.239  0.21550    
monthMay             -7.163e-04  2.485e-03  -0.288  0.77320    
monthNis             -3.008e-03  2.612e-03  -1.152  0.24980    
monthOca             -3.981e-03  3.234e-03  -1.231  0.21862    
monthŞub             -4.200e-03  3.328e-03  -1.262  0.20728    
monthTem              5.044e-03  2.262e-03   2.230  0.02601 *  
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 0.0119 on 847 degrees of freedom
Multiple R-squared:  0.6806,	Adjusted R-squared:  0.6742 
F-statistic: 106.2 on 17 and 847 DF,  p-value: < 2.2e-16

	Breusch-Godfrey test for serial correlation of order up to 21

data:  Residuals
LM test = 415.11, df = 21, p-value < 2.2e-16


#Production data for hour 4
hour_4_data <- all_data[all_data$hour == 4, ]

# Plot production for hour 4
ggplot(hour_4_data, aes(x = date, y = production)) +
  geom_line() +
  labs(title = "Hourly Production Data for Hour 4",
       x = "Date",
       y = "Production") +
  theme_minimal()


hour_4_data <- all_data %>%
  filter(hour == 4, date >= start_date & date <= end_date)

forecast_data_hour_4 <- to_be_forecasted %>%
  filter(hour == 4)
forecast_data_hour_4$trend_hour_4 <- nrow(hour_4_data) + 1:nrow(forecast_data_hour_4)

last_production_value <- tail(hour_4_data$production, 1)
forecast_data_hour_4 <- forecast_data_hour_4 %>%
  mutate(lag_1_production = last_production_value)

predictions_hour_4 <- predict(lm_hour_4, newdata = forecast_data_hour_4)
predictions_hour_4[predictions_hour_4 < 0] <- 0

predictions_hour_4_df <- data.frame(
  date = forecast_data_hour_4$date,  
  predicted = predictions_hour_4
)

merged_data <- merge(hour_4_data, predictions_hour_4_df, by = "date")

output_data <- merged_data %>%
  select(date, actual = production, predicted)

output_data <- output_data %>%
  mutate(error = abs(predicted - actual))

wmape <- sum(output_data$error) / sum(output_data$actual) * 100

print(paste("WMAPE:", wmape, "%"))

output_dir <- "C:/Users/ecemozturk/Desktop"
if (!dir.exists(output_dir)) {
  dir.create(output_dir, recursive = TRUE)
}

file_path <- file.path(output_dir, "predictions_hour_4_with_actuals.xlsx")

write_xlsx(output_data, file_path)

print(output_data)

[1] "WMAPE: 1407.11310958475 %"
Key: <date>
           date actual  predicted       error
         <IDat>  <num>      <num>       <num>
  1: 2024-02-01   0.00 0.05524756 0.055247562
  2: 2024-02-02   0.00 0.05539115 0.055391154
  3: 2024-02-03   0.00 0.05517195 0.055171948
  4: 2024-02-04   0.00 0.05462935 0.054629350
  5: 2024-02-05   0.00 0.05435240 0.054352396
 ---                                         
101: 2024-05-11   0.03 0.05726587 0.027265872
102: 2024-05-12   0.00 0.05773654 0.057736539
103: 2024-05-13   0.06 0.05737257 0.002627426
104: 2024-05-14   0.07 0.05818364 0.011816364
105: 2024-05-15   0.08 0.05746433 0.022535670


# Filter data for hour 5
hour_5_data <- all_data[all_data$hour == 5, ]
hour_5_data <- hour_5_data[,-c(2)]
hour_5_data[, lag_1_production := shift(production,1)]
hour_5_data[,lag_1_diff:=production-lag_1_production]
hour_5_data <- hour_5_data[!is.na(lag_1_production)]
hour_5_data$trend_hour_5 <- 1:nrow(hour_5_data)
# Fit linear regression model for hour 5
lm_hour_5 <- lm(production ~+lag_1_production+dlwrf_surface+is.ramadan+special_period+trend_hour_5 +month*hourly_max_t, data = hour_5_data)
# Summarize the model
summary(lm_hour_5)
checkresiduals(lm_hour_5)
plot(lm_hour_5)

Call:
lm(formula = production ~ +lag_1_production + dlwrf_surface + 
    is.ramadan + special_period + trend_hour_5 + month * hourly_max_t, 
    data = hour_5_data)

Residuals:
     Min       1Q   Median       3Q      Max 
-0.97514 -0.03869 -0.00864  0.02658  1.76482 

Coefficients:
                        Estimate Std. Error t value Pr(>|t|)    
(Intercept)           -1.507e+00  2.805e+00  -0.537   0.5912    
lag_1_production       5.048e-01  3.000e-02  16.826  < 2e-16 ***
dlwrf_surface          3.991e-04  2.609e-04   1.530   0.1265    
is.ramadan            -4.754e-02  2.091e-02  -2.273   0.0233 *  
special_period        -7.987e-02  1.669e-02  -4.785 2.02e-06 ***
trend_hour_5           1.222e-04  2.369e-05   5.157 3.13e-07 ***
monthAra               2.776e+00  3.181e+00   0.873   0.3830    
monthEki               3.291e+00  3.214e+00   1.024   0.3061    
monthEyl               2.167e+00  3.253e+00   0.666   0.5054    
monthHaz              -5.195e+00  4.423e+00  -1.175   0.2405    
monthKas               3.170e+00  3.191e+00   0.994   0.3207    
monthMar               2.567e+00  2.860e+00   0.898   0.3697    
monthMay               4.199e+00  3.186e+00   1.318   0.1879    
monthNis              -8.816e-01  3.015e+00  -0.292   0.7701    
monthOca               3.002e+00  2.879e+00   1.043   0.2974    
monthŞub               2.648e+00  2.868e+00   0.923   0.3561    
monthTem              -2.084e+00  3.595e+00  -0.580   0.5623    
hourly_max_t           5.028e-03  9.818e-03   0.512   0.6087    
monthAra:hourly_max_t -1.029e-02  1.122e-02  -0.917   0.3595    
monthEki:hourly_max_t -1.184e-02  1.127e-02  -1.051   0.2936    
monthEyl:hourly_max_t -7.717e-03  1.138e-02  -0.678   0.4977    
monthHaz:hourly_max_t  1.857e-02  1.551e-02   1.197   0.2315    
monthKas:hourly_max_t -1.158e-02  1.123e-02  -1.031   0.3027    
monthMar:hourly_max_t -9.429e-03  1.000e-02  -0.943   0.3460    
monthMay:hourly_max_t -1.502e-02  1.117e-02  -1.345   0.1791    
monthNis:hourly_max_t  3.273e-03  1.055e-02   0.310   0.7565    
monthOca:hourly_max_t -1.111e-02  1.008e-02  -1.103   0.2706    
monthŞub:hourly_max_t -9.825e-03  1.004e-02  -0.979   0.3280    
monthTem:hourly_max_t  7.583e-03  1.256e-02   0.604   0.5463    
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 0.1276 on 836 degrees of freedom
Multiple R-squared:  0.6002,	Adjusted R-squared:  0.5868 
F-statistic: 44.83 on 28 and 836 DF,  p-value: < 2.2e-16

	Breusch-Godfrey test for serial correlation of order up to 32

data:  Residuals
LM test = 247.14, df = 32, p-value < 2.2e-16


#Production data for hour 5
hour_5_data <- all_data[all_data$hour == 5, ]

# Plot production for hour 4
ggplot(hour_5_data, aes(x = date, y = production)) +
  geom_line() +
  labs(title = "Hourly Production Data for Hour 5",
       x = "Date",
       y = "Production") +
  theme_minimal()


hour_5_data <- all_data %>%
  filter(hour == 5, date >= start_date & date <= end_date)

forecast_data_hour_5 <- to_be_forecasted %>%
  filter(hour == 5)
forecast_data_hour_5$trend_hour_5 <- nrow(hour_5_data) + 1:nrow(forecast_data_hour_5)

last_production_value <- tail(hour_5_data$production, 1)
forecast_data_hour_5 <- forecast_data_hour_5 %>%
  mutate(lag_1_production = last_production_value)

predictions_hour_5 <- predict(lm_hour_5, newdata = forecast_data_hour_5)
predictions_hour_5[predictions_hour_5 < 0] <- 0

predictions_hour_5_df <- data.frame(
  date = forecast_data_hour_5$date,  
  predicted = predictions_hour_5
)

merged_data <- merge(hour_5_data, predictions_hour_5_df, by = "date")

output_data <- merged_data %>%
  select(date, actual = production, predicted)

output_data <- output_data %>%
  mutate(error = abs(predicted - actual))

wmape <- sum(output_data$error) / sum(output_data$actual) * 100

print(paste("WMAPE:", wmape, "%"))

output_dir <- "C:/Users/ecemozturk/Desktop"
if (!dir.exists(output_dir)) {
  dir.create(output_dir, recursive = TRUE)
}

file_path <- file.path(output_dir, "predictions_hour_5_with_actuals.xlsx")

write_xlsx(output_data, file_path)

print(output_data)

[1] "WMAPE: 154.103457098309 %"
Key: <date>
           date actual predicted     error
         <IDat>  <num>     <num>     <num>
  1: 2024-02-01   0.00 0.3766761 0.3766761
  2: 2024-02-02   0.00 0.3754115 0.3754115
  3: 2024-02-03   0.00 0.3770272 0.3770272
  4: 2024-02-04   0.00 0.3780127 0.3780127
  5: 2024-02-05   0.00 0.3854003 0.3854003
 ---                                      
101: 2024-05-11   0.64 0.4225375 0.2174625
102: 2024-05-12   0.00 0.4543419 0.4543419
103: 2024-05-13   0.76 0.4710763 0.2889237
104: 2024-05-14   0.96 0.4800365 0.4799635
105: 2024-05-15   0.84 0.4798892 0.3601108


# Filter data for hour 6
hour_6_data <- all_data[all_data$hour == 6, ]
hour_6_data <- hour_6_data[, -c(2)]
# Convert the data frame to a data.table
setDT(hour_6_data)
# Create a lagged variable for production with a lag of 1 period
hour_6_data[, lag_1_production := shift(production,1)]
hour_6_data[,lag_1_diff:=production-lag_1_production]
hour_6_data <- hour_6_data[!is.na(lag_1_production)]
hour_6_data$trend_hour_6 <- 1:nrow(hour_6_data)
# Fit linear regression model for hour 6
lm_hour_6 <- lm(production~+lag_1_production+uswrf_top_of_atmosphere+wday+ hourly_cloud_average+is.ramadan+special_period+is.religousday+month*hourly_max_t, data = hour_6_data) 
#lm_hour_6 <- lm(production ~ +uswrf_top_of_atmosphere + is.ramadan +is.weekend+ is.religousday +is.publicholiday + month*hourly_max_t , data = hour_6_data)
# Summarize the model
summary(lm_hour_6)
checkresiduals(lm_hour_6)
plot(lm_hour_6)

Call:
lm(formula = production ~ +lag_1_production + uswrf_top_of_atmosphere + 
    wday + hourly_cloud_average + is.ramadan + special_period + 
    is.religousday + month * hourly_max_t, data = hour_6_data)

Residuals:
    Min      1Q  Median      3Q     Max 
-2.5222 -0.2081 -0.0312  0.1482  4.4413 

Coefficients:
                          Estimate Std. Error t value Pr(>|t|)    
(Intercept)             -1.368e+00  1.318e+01  -0.104   0.9173    
lag_1_production         4.266e-01  3.218e-02  13.256   <2e-16 ***
uswrf_top_of_atmosphere  5.519e-02  3.712e-02   1.487   0.1375    
wdayCum                 -1.306e-01  7.841e-02  -1.666   0.0962 .  
wdayÇar                  4.254e-02  7.888e-02   0.539   0.5898    
wdayPaz                 -1.408e-01  7.828e-02  -1.799   0.0724 .  
wdayPer                 -7.452e-02  7.865e-02  -0.947   0.3437    
wdayPts                 -3.051e-02  7.888e-02  -0.387   0.6990    
wdaySal                 -1.332e-01  7.870e-02  -1.693   0.0909 .  
hourly_cloud_average    -2.392e-03  9.442e-04  -2.533   0.0115 *  
is.ramadan              -1.829e-01  9.955e-02  -1.837   0.0666 .  
special_period          -7.515e-01  7.904e-02  -9.508   <2e-16 ***
is.religousday           1.018e-01  1.330e-01   0.766   0.4440    
monthAra                -3.779e+00  1.518e+01  -0.249   0.8035    
monthEki                 9.220e+00  1.524e+01   0.605   0.5453    
monthEyl                 4.213e-01  1.545e+01   0.027   0.9783    
monthHaz                 9.651e+00  2.127e+01   0.454   0.6502    
monthKas                 3.229e+00  1.518e+01   0.213   0.8316    
monthMar                -1.643e+00  1.354e+01  -0.121   0.9035    
monthMay                 2.190e+01  1.520e+01   1.441   0.1500    
monthNis                -1.766e+01  1.435e+01  -1.231   0.2187    
monthOca                -3.706e-02  1.362e+01  -0.003   0.9978    
monthŞub                -1.313e-01  1.358e+01  -0.010   0.9923    
monthTem                 1.194e+00  1.731e+01   0.069   0.9450    
hourly_max_t             8.200e-03  4.598e-02   0.178   0.8585    
monthAra:hourly_max_t    1.123e-02  5.364e-02   0.209   0.8343    
monthEki:hourly_max_t   -3.356e-02  5.350e-02  -0.627   0.5306    
monthEyl:hourly_max_t   -1.327e-03  5.413e-02  -0.025   0.9804    
monthHaz:hourly_max_t   -3.341e-02  7.440e-02  -0.449   0.6534    
monthKas:hourly_max_t   -1.354e-02  5.348e-02  -0.253   0.8002    
monthMar:hourly_max_t    4.398e-03  4.739e-02   0.093   0.9261    
monthMay:hourly_max_t   -7.962e-02  5.342e-02  -1.491   0.1365    
monthNis:hourly_max_t    6.322e-02  5.027e-02   1.258   0.2089    
monthOca:hourly_max_t   -2.352e-03  4.774e-02  -0.049   0.9607    
monthŞub:hourly_max_t   -1.929e-03  4.758e-02  -0.041   0.9677    
monthTem:hourly_max_t   -2.952e-03  6.055e-02  -0.049   0.9611    
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 0.6133 on 829 degrees of freedom
Multiple R-squared:  0.6422,	Adjusted R-squared:  0.6271 
F-statistic: 42.52 on 35 and 829 DF,  p-value: < 2.2e-16

	Breusch-Godfrey test for serial correlation of order up to 39

data:  Residuals
LM test = 196.53, df = 39, p-value < 2.2e-16


hour_6_data <- all_data[all_data$hour == 6, ]

# Plot production for hour 6
ggplot(hour_6_data, aes(x = date, y = production)) +
  geom_line() +
  labs(title = "Hourly Production Data for Hour 6",
       x = "Date",
       y = "Production") +
  theme_minimal()


hour_6_data <- all_data %>%
  filter(hour == 6, date >= start_date & date <= end_date)

forecast_data_hour_6 <- to_be_forecasted %>%
  filter(hour == 6)
forecast_data_hour_6$trend_hour_6 <- nrow(hour_6_data) + 1:nrow(forecast_data_hour_6)

last_production_value <- tail(hour_6_data$production, 1)
forecast_data_hour_6 <- forecast_data_hour_6 %>%
  mutate(lag_1_production = last_production_value)

predictions_hour_6 <- predict(lm_hour_6, newdata = forecast_data_hour_6)
predictions_hour_6[predictions_hour_6 < 0] <- 0

predictions_hour_6_df <- data.frame(
  date = forecast_data_hour_6$date,  
  predicted = predictions_hour_6
)

merged_data <- merge(hour_6_data, predictions_hour_6_df, by = "date")

output_data <- merged_data %>%
  select(date, actual = production, predicted)

output_data <- output_data %>%
  mutate(error = abs(predicted - actual))

wmape <- sum(output_data$error) / sum(output_data$actual) * 100

print(paste("WMAPE:", wmape, "%"))

output_dir <- "C:/Users/ecemozturk/Desktop"
if (!dir.exists(output_dir)) {
  dir.create(output_dir, recursive = TRUE)
}

file_path <- file.path(output_dir, "predictions_hour_6_with_actuals.xlsx")

write_xlsx(output_data, file_path)

print(output_data)

[1] "WMAPE: 66.2971330635574 %"
Key: <date>
           date actual predicted     error
         <IDat>  <num>     <num>     <num>
  1: 2024-02-01   0.01  1.302795 1.2927954
  2: 2024-02-02   0.01  1.290989 1.2809892
  3: 2024-02-03   0.01  1.375376 1.3653757
  4: 2024-02-04   0.07  1.148062 1.0780617
  5: 2024-02-05   0.00  1.209222 1.2092218
 ---                                      
101: 2024-05-11   2.51  1.661629 0.8483713
102: 2024-05-12   0.00  1.599317 1.5993171
103: 2024-05-13   2.69  1.814346 0.8756536
104: 2024-05-14   2.59  1.908270 0.6817298
105: 2024-05-15   2.94  2.044396 0.8956045


library(data.table)

# Filter data for hour 7 and remove the hour column
hour_7_data <- all_data[all_data$hour == 7, ]
hour_7_data <- hour_7_data[, -c(2)]

# Create a trend variable for hour 7
hour_7_data$trend_hour_7 <- 1:nrow(hour_7_data)

# Convert the data frame to a data.table
setDT(hour_7_data)

# Create a lagged variable for production with a lag of 1 period
hour_7_data[, lag_1_production := shift(production,1)]
hour_7_data[,lag_1_diff:=production-lag_1_production]
# Remove rows with NA in lagged production to ensure the model can run
hour_7_data <- hour_7_data[!is.na(lag_1_production)]

# Fit linear regression model for hour 7 including the lagged variable
lm_hour_7 <- lm(production ~+lag_1_production+trend_hour_7 + special_period + dlwrf_surface + hourly_cloud_average + month+hourly_max_t,data=hour_7_data)
summary(lm_hour_7)

# Check residuals
library(forecast)
checkresiduals(lm_hour_7)

# Plot the model
plot(lm_hour_7)

Call:
lm(formula = production ~ +lag_1_production + trend_hour_7 + 
    special_period + dlwrf_surface + hourly_cloud_average + month + 
    hourly_max_t, data = hour_7_data)

Residuals:
    Min      1Q  Median      3Q     Max 
-5.4844 -0.6254  0.0878  0.7845  9.0720 

Coefficients:
                       Estimate Std. Error t value Pr(>|t|)    
(Intercept)          -2.009e+01  5.698e+00  -3.526 0.000444 ***
lag_1_production      2.783e-01  3.120e-02   8.918  < 2e-16 ***
trend_hour_7          6.333e-04  2.351e-04   2.693 0.007215 ** 
special_period       -5.767e-01  1.690e-01  -3.413 0.000673 ***
dlwrf_surface        -1.845e-02  4.190e-03  -4.402 1.21e-05 ***
hourly_cloud_average -8.618e-03  3.471e-03  -2.483 0.013218 *  
monthAra             -1.474e+00  3.846e-01  -3.834 0.000135 ***
monthEki              7.961e-01  3.071e-01   2.592 0.009700 ** 
monthEyl              7.637e-01  2.710e-01   2.819 0.004937 ** 
monthHaz              1.155e+00  2.637e-01   4.378 1.35e-05 ***
monthKas             -5.118e-01  3.473e-01  -1.474 0.140955    
monthMar              1.082e-01  3.813e-01   0.284 0.776751    
monthMay              5.781e-01  2.890e-01   2.001 0.045751 *  
monthNis              6.890e-01  3.066e-01   2.247 0.024893 *  
monthOca             -1.459e+00  4.083e-01  -3.575 0.000370 ***
monthŞub             -6.433e-01  4.173e-01  -1.542 0.123544    
monthTem              1.014e+00  2.601e-01   3.901 0.000104 ***
hourly_max_t          9.756e-02  2.323e-02   4.200 2.95e-05 ***
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 1.381 on 847 degrees of freedom
Multiple R-squared:  0.6439,	Adjusted R-squared:  0.6367 
F-statistic: 90.08 on 17 and 847 DF,  p-value: < 2.2e-16

	Breusch-Godfrey test for serial correlation of order up to 21

data:  Residuals
LM test = 65.049, df = 21, p-value = 2.136e-06


# Plot production for hour 7
ggplot(hour_7_data, aes(x = date, y = production)) +
  geom_line() +
  labs(title = "Hourly Production Data for Hour 7",
       x = "Date",
       y = "Production") +
  theme_minimal()

Warning message in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <c5>”
Warning message in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <9e>”
Warning message in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <c5>”
Warning message in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <9e>”


hour_7_data <- all_data %>%
  filter(hour == 7, date >= start_date & date <= end_date)

forecast_data_hour_7 <- to_be_forecasted %>%
  filter(hour == 7)
forecast_data_hour_7$trend_hour_7 <- nrow(hour_7_data) + 1:nrow(forecast_data_hour_7)

last_production_value <- tail(hour_7_data$production, 1)
forecast_data_hour_7 <- forecast_data_hour_7 %>%
  mutate(lag_1_production = last_production_value)

predictions_hour_7 <- predict(lm_hour_7, newdata = forecast_data_hour_7)
predictions_hour_7[predictions_hour_7 < 0] <- 0

predictions_hour_7_df <- data.frame(
  date = forecast_data_hour_7$date,  
  predicted = predictions_hour_7
)

merged_data <- merge(hour_7_data, predictions_hour_7_df, by = "date")

output_data <- merged_data %>%
  select(date, actual = production, predicted)

output_data <- output_data %>%
  mutate(error = abs(predicted - actual))

wmape <- sum(output_data$error) / sum(output_data$actual) * 100

print(paste("WMAPE:", wmape, "%"))

output_dir <- "C:/Users/ecemozturk/Desktop"
if (!dir.exists(output_dir)) {
  dir.create(output_dir, recursive = TRUE)
}

file_path <- file.path(output_dir, "predictions_hour_7_with_actuals.xlsx")

write_xlsx(output_data, file_path)

print(output_data)

[1] "WMAPE: 29.7652596829167 %"
Key: <date>
           date actual predicted     error
         <IDat>  <num>     <num>     <num>
  1: 2024-02-01   1.59  3.143221 1.5532210
  2: 2024-02-02   1.62  3.556651 1.9366508
  3: 2024-02-03   0.94  3.054084 2.1140840
  4: 2024-02-04   1.03  2.487699 1.4576990
  5: 2024-02-05   0.88  1.745296 0.8652960
 ---                                      
101: 2024-05-11   6.27  4.955730 1.3142697
102: 2024-05-12   0.00  3.903619 3.9036191
103: 2024-05-13   5.75  3.460671 2.2893287
104: 2024-05-14   4.14  3.880553 0.2594466
105: 2024-05-15   6.58  4.327481 2.2525187


# Filter data for hour 8
hour_8_data <- all_data[all_data$hour == 8, ]
hour_8_data <- hour_8_data[,-c(2)]
hour_8_data$trend_hour_8 <- 1:nrow(hour_8_data)
# Create a lagged variable for production with a lag of 1 period
hour_8_data[, lag_1_production := shift(production,1)]
hour_8_data[,lag_1_diff:=production-lag_1_production]
# Remove rows with NA in lagged production to ensure the model can run
hour_8_data <- hour_8_data[!is.na(lag_1_production)]
#Fit linear regression model for hour 8
lm_hour_8 <- lm(production ~ +lag_1_production+trend_hour_8+ csnow_surface+dlwrf_surface+hourly_cloud_average+is.ramadan+month*hourly_max_t, data = hour_8_data)
# Summarize the model
summary(lm_hour_8)

checkresiduals(lm_hour_8)
plot(lm_hour_8)

Call:
lm(formula = production ~ +lag_1_production + trend_hour_8 + 
    csnow_surface + dlwrf_surface + hourly_cloud_average + is.ramadan + 
    month * hourly_max_t, data = hour_8_data)

Residuals:
    Min      1Q  Median      3Q     Max 
-7.7347 -0.9508  0.2114  1.0988  5.7171 

Coefficients:
                        Estimate Std. Error t value Pr(>|t|)    
(Intercept)            3.007e+00  3.530e+01   0.085  0.93214    
lag_1_production       1.645e-01  2.916e-02   5.640 2.33e-08 ***
trend_hour_8           1.089e-03  2.785e-04   3.910 9.96e-05 ***
csnow_surface         -1.217e+00  4.449e-01  -2.736  0.00636 ** 
dlwrf_surface         -2.800e-02  5.213e-03  -5.373 1.01e-07 ***
hourly_cloud_average  -2.446e-02  4.726e-03  -5.174 2.86e-07 ***
is.ramadan            -6.163e-01  3.035e-01  -2.031  0.04259 *  
monthAra              -6.920e+01  4.174e+01  -1.658  0.09774 .  
monthEki              -2.122e+01  4.112e+01  -0.516  0.60601    
monthEyl               2.384e+01  4.017e+01   0.593  0.55305    
monthHaz              -1.170e+01  4.634e+01  -0.252  0.80072    
monthKas              -2.417e+01  4.048e+01  -0.597  0.55073    
monthMar              -3.480e+01  3.654e+01  -0.952  0.34121    
monthMay              -5.104e+00  3.892e+01  -0.131  0.89569    
monthNis              -1.825e+01  3.752e+01  -0.486  0.62684    
monthOca              -4.946e+01  3.652e+01  -1.354  0.17605    
monthŞub              -3.804e+01  3.634e+01  -1.047  0.29550    
monthTem              -1.275e+01  4.553e+01  -0.280  0.77950    
hourly_max_t           3.750e-02  1.191e-01   0.315  0.75291    
monthAra:hourly_max_t  2.434e-01  1.438e-01   1.692  0.09107 .  
monthEki:hourly_max_t  7.415e-02  1.398e-01   0.531  0.59584    
monthEyl:hourly_max_t -8.233e-02  1.353e-01  -0.608  0.54312    
monthHaz:hourly_max_t  4.065e-02  1.564e-01   0.260  0.79500    
monthKas:hourly_max_t  8.094e-02  1.383e-01   0.585  0.55858    
monthMar:hourly_max_t  1.242e-01  1.233e-01   1.007  0.31409    
monthMay:hourly_max_t  1.699e-02  1.313e-01   0.129  0.89707    
monthNis:hourly_max_t  6.320e-02  1.264e-01   0.500  0.61721    
monthOca:hourly_max_t  1.710e-01  1.235e-01   1.385  0.16638    
monthŞub:hourly_max_t  1.333e-01  1.227e-01   1.086  0.27767    
monthTem:hourly_max_t  4.460e-02  1.531e-01   0.291  0.77096    
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 1.818 on 835 degrees of freedom
Multiple R-squared:  0.5892,	Adjusted R-squared:  0.5749 
F-statistic: 41.29 on 29 and 835 DF,  p-value: < 2.2e-16

	Breusch-Godfrey test for serial correlation of order up to 33

data:  Residuals
LM test = 68.172, df = 33, p-value = 0.0003044


# Plot production for hour 8
ggplot(hour_8_data, aes(x = date, y = production)) +
  geom_line() +
  labs(title = "Hourly Production Data for Hour 8",
       x = "Date",
       y = "Production") +
  theme_minimal()


hour_8_data <- all_data %>%
  filter(hour == 8, date >= start_date & date <= end_date)

forecast_data_hour_8 <- to_be_forecasted %>%
  filter(hour == 8)
forecast_data_hour_8$trend_hour_8 <- nrow(hour_8_data) + 1:nrow(forecast_data_hour_8)

last_production_value <- tail(hour_8_data$production, 1)
forecast_data_hour_8 <- forecast_data_hour_8 %>%
  mutate(lag_1_production = last_production_value)

predictions_hour_8 <- predict(lm_hour_8, newdata = forecast_data_hour_8)
predictions_hour_8[predictions_hour_8 < 0] <- 0

predictions_hour_8_df <- data.frame(
  date = forecast_data_hour_8$date,  
  predicted = predictions_hour_8
)

merged_data <- merge(hour_8_data, predictions_hour_8_df, by = "date")

output_data <- merged_data %>%
  select(date, actual = production, predicted)

output_data <- output_data %>%
  mutate(error = abs(predicted - actual))

wmape <- sum(output_data$error) / sum(output_data$actual) * 100

print(paste("WMAPE:", wmape, "%"))

output_dir <- "C:/Users/ecemozturk/Desktop"
if (!dir.exists(output_dir)) {
  dir.create(output_dir, recursive = TRUE)
}

file_path <- file.path(output_dir, "predictions_hour_8_with_actuals.xlsx")

write_xlsx(output_data, file_path)

print(output_data)

[1] "WMAPE: 22.6917817588993 %"
Key: <date>
           date actual predicted     error
         <IDat>  <num>     <num>     <num>
  1: 2024-02-01   6.17  5.694100 0.4759005
  2: 2024-02-02   6.15  6.627810 0.4778098
  3: 2024-02-03   4.21  5.471073 1.2610726
  4: 2024-02-04   5.04  4.426163 0.6138374
  5: 2024-02-05   3.01  2.770403 0.2395970
 ---                                      
101: 2024-05-11   9.07  7.128347 1.9416530
102: 2024-05-12   0.00  5.323364 5.3233640
103: 2024-05-13   9.24  4.310057 4.9299429
104: 2024-05-14   7.60  5.473529 2.1264705
105: 2024-05-15   9.13  5.993096 3.1369043


# Filter data for hour 9
hour_9_data <- all_data[all_data$hour == 9, ]
hour_9_data <- hour_9_data[,-c(2)]
hour_9_data$trend_hour_9 <- 1:nrow(hour_9_data)
hour_9_data[, lag_1_production := shift(production,1)]
hour_9_data[,lag_1_diff:=production-lag_1_production]
# Remove rows with NA in lagged production to ensure the model can run
hour_9_data <- hour_9_data[!is.na(lag_1_production)]
# Fit linear regression model for hour 9
lm_hour_9 <- lm(production ~+lag_1_production+special_period+dswrf_surface+csnow_surface+dlwrf_surface+is.nationalday+uswrf_surface+hourly_cloud_average+month*hourly_max_t, data = hour_9_data)
# Summarize the model
summary(lm_hour_9)
checkresiduals(lm_hour_9)
plot(lm_hour_9)

Call:
lm(formula = production ~ +lag_1_production + special_period + 
    dswrf_surface + csnow_surface + dlwrf_surface + is.nationalday + 
    uswrf_surface + hourly_cloud_average + month * hourly_max_t, 
    data = hour_9_data)

Residuals:
    Min      1Q  Median      3Q     Max 
-8.3796 -1.0059  0.2701  1.2028  5.2072 

Coefficients:
                        Estimate Std. Error t value Pr(>|t|)    
(Intercept)           -2.673e+01  3.908e+01  -0.684  0.49423    
lag_1_production       8.335e-02  2.831e-02   2.945  0.00332 ** 
special_period        -3.158e-01  2.134e-01  -1.480  0.13923    
dswrf_surface         -3.688e-02  1.474e-02  -2.502  0.01255 *  
csnow_surface         -1.383e+00  5.236e-01  -2.641  0.00841 ** 
dlwrf_surface         -3.795e-02  6.159e-03  -6.161 1.12e-09 ***
is.nationalday         1.015e+00  5.652e-01   1.796  0.07284 .  
uswrf_surface          1.028e-01  4.429e-02   2.320  0.02058 *  
hourly_cloud_average  -2.812e-02  5.303e-03  -5.304 1.45e-07 ***
monthAra              -1.102e+02  4.719e+01  -2.334  0.01982 *  
monthEki              -3.741e+01  4.523e+01  -0.827  0.40833    
monthEyl               2.276e+01  4.286e+01   0.531  0.59555    
monthHaz              -2.816e+01  4.784e+01  -0.589  0.55618    
monthKas              -1.590e+01  4.274e+01  -0.372  0.70990    
monthMar              -3.979e+01  4.214e+01  -0.944  0.34530    
monthMay              -1.751e+01  4.198e+01  -0.417  0.67663    
monthNis              -1.645e+01  4.081e+01  -0.403  0.68701    
monthOca              -7.982e+01  4.042e+01  -1.975  0.04860 *  
monthŞub              -2.391e+01  4.025e+01  -0.594  0.55266    
monthTem               2.289e+01  4.695e+01   0.488  0.62600    
hourly_max_t           1.581e-01  1.299e-01   1.217  0.22396    
monthAra:hourly_max_t  3.981e-01  1.604e-01   2.481  0.01329 *  
monthEki:hourly_max_t  1.316e-01  1.506e-01   0.874  0.38220    
monthEyl:hourly_max_t -7.465e-02  1.411e-01  -0.529  0.59699    
monthHaz:hourly_max_t  1.012e-01  1.584e-01   0.639  0.52314    
monthKas:hourly_max_t  5.577e-02  1.424e-01   0.392  0.69550    
monthMar:hourly_max_t  1.448e-01  1.402e-01   1.032  0.30227    
monthMay:hourly_max_t  6.412e-02  1.388e-01   0.462  0.64420    
monthNis:hourly_max_t  5.935e-02  1.347e-01   0.441  0.65955    
monthOca:hourly_max_t  2.893e-01  1.342e-01   2.156  0.03139 *  
monthŞub:hourly_max_t  8.869e-02  1.334e-01   0.665  0.50621    
monthTem:hourly_max_t -7.121e-02  1.546e-01  -0.461  0.64511    
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 1.981 on 833 degrees of freedom
Multiple R-squared:  0.5639,	Adjusted R-squared:  0.5476 
F-statistic: 34.74 on 31 and 833 DF,  p-value: < 2.2e-16

	Breusch-Godfrey test for serial correlation of order up to 35

data:  Residuals
LM test = 20.373, df = 35, p-value = 0.9769


# Plot production for hour 9
ggplot(hour_9_data, aes(x = date, y = production)) +
  geom_line() +
  labs(title = "Hourly Production Data for Hour 9",
       x = "Date",
       y = "Production") +
  theme_minimal()

Warning message in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <c5>”
Warning message in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <9e>”
Warning message in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <c5>”
Warning message in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <9e>”


hour_9_data <- all_data %>%
  filter(hour == 9, date >= start_date & date <= end_date)

forecast_data_hour_9 <- to_be_forecasted %>%
  filter(hour == 9)
forecast_data_hour_9$trend_hour_9 <- nrow(hour_9_data) + 1:nrow(forecast_data_hour_9)

last_production_value <- tail(hour_9_data$production, 1)
forecast_data_hour_9 <- forecast_data_hour_9 %>%
  mutate(lag_1_production = last_production_value)

predictions_hour_9 <- predict(lm_hour_9, newdata = forecast_data_hour_9)
predictions_hour_9[predictions_hour_9 < 0] <- 0

predictions_hour_9_df <- data.frame(
  date = forecast_data_hour_9$date,  
  predicted = predictions_hour_9
)

merged_data <- merge(hour_9_data, predictions_hour_9_df, by = "date")

output_data <- merged_data %>%
  select(date, actual = production, predicted)

output_data <- output_data %>%
  mutate(error = abs(predicted - actual))

wmape <- sum(output_data$error) / sum(output_data$actual) * 100

print(paste("WMAPE:", wmape, "%"))

output_dir <- "C:/Users/ecemozturk/Desktop"
if (!dir.exists(output_dir)) {
  dir.create(output_dir, recursive = TRUE)
}

file_path <- file.path(output_dir, "predictions_hour_9_with_actuals.xlsx")

write_xlsx(output_data, file_path)

print(output_data)

[1] "WMAPE: 19.5185218600221 %"
Key: <date>
           date actual predicted     error
         <IDat>  <num>     <num>     <num>
  1: 2024-02-01   7.11  8.060603 0.9506034
  2: 2024-02-02   9.03  9.322309 0.2923085
  3: 2024-02-03   6.83  7.574831 0.7448312
  4: 2024-02-04   9.02  6.479980 2.5400196
  5: 2024-02-05   4.21  4.660045 0.4500446
 ---                                      
101: 2024-05-11   9.85  8.116059 1.7339413
102: 2024-05-12   0.00  6.053854 6.0538535
103: 2024-05-13   8.88  4.573623 4.3063768
104: 2024-05-14   8.83  5.654152 3.1758482
105: 2024-05-15   9.82  6.814517 3.0054834


# Create a trend variable for hour 10
hour_10_data <- all_data[all_data$hour == 10, ]
hour_10_data <- hour_10_data[,-c(2)]
hour_10_data$trend_hour_10 <- 1:nrow(hour_10_data)
hour_10_data[, lag_1_production := shift(production,1)]
hour_10_data[,lag_1_diff:=production-lag_1_production]
# Remove rows with NA in lagged production to ensure the model can run
hour_10_data <- hour_10_data[!is.na(lag_1_production)]
# Fit linear regression model for hour 10
lm_hour_10 <- lm(production ~+lag_1_production +csnow_surface+dlwrf_surface+hourly_cloud_average+is.weekend+is.religousday+is.nationalday+month*hourly_max_t, data = hour_10_data)
summary(lm_hour_10)

checkresiduals(lm_hour_10)
plot(lm_hour_10)

pacf(hour_10_data$production)

Call:
lm(formula = production ~ +lag_1_production + csnow_surface + 
    dlwrf_surface + hourly_cloud_average + is.weekend + is.religousday + 
    is.nationalday + month * hourly_max_t, data = hour_10_data)

Residuals:
    Min      1Q  Median      3Q     Max 
-9.8962 -0.6891  0.2512  1.0714  5.4454 

Coefficients:
                        Estimate Std. Error t value Pr(>|t|)    
(Intercept)           -2.762e+01  3.911e+01  -0.706  0.48024    
lag_1_production       7.410e-02  2.789e-02   2.657  0.00804 ** 
csnow_surface         -2.278e+00  5.748e-01  -3.964 8.01e-05 ***
dlwrf_surface         -3.159e-02  4.756e-03  -6.643 5.55e-11 ***
hourly_cloud_average  -2.114e-02  5.198e-03  -4.067 5.21e-05 ***
is.weekend            -1.860e-01  1.552e-01  -1.198  0.23113    
is.religousday         8.104e-01  4.378e-01   1.851  0.06450 .  
is.nationalday         7.260e-01  5.847e-01   1.242  0.21470    
monthAra              -1.330e+02  4.784e+01  -2.780  0.00556 ** 
monthEki              -2.118e+01  4.538e+01  -0.467  0.64071    
monthEyl               3.466e+00  4.299e+01   0.081  0.93575    
monthHaz              -4.052e+01  4.656e+01  -0.870  0.38434    
monthKas              -1.036e+01  4.223e+01  -0.245  0.80617    
monthMar               1.790e+01  4.028e+01   0.445  0.65679    
monthMay              -3.094e+00  4.168e+01  -0.074  0.94084    
monthNis               3.873e+00  4.079e+01   0.095  0.92437    
monthOca              -7.934e+01  4.127e+01  -1.922  0.05490 .  
monthŞub               1.043e+01  4.058e+01   0.257  0.79716    
monthTem               1.904e+01  4.619e+01   0.412  0.68038    
hourly_max_t           1.517e-01  1.266e-01   1.199  0.23102    
monthAra:hourly_max_t  4.782e-01  1.600e-01   2.989  0.00288 ** 
monthEki:hourly_max_t  7.438e-02  1.483e-01   0.501  0.61616    
monthEyl:hourly_max_t -1.009e-02  1.390e-01  -0.073  0.94216    
monthHaz:hourly_max_t  1.358e-01  1.514e-01   0.897  0.36997    
monthKas:hourly_max_t  3.748e-02  1.378e-01   0.272  0.78568    
monthMar:hourly_max_t -5.926e-02  1.306e-01  -0.454  0.65010    
monthMay:hourly_max_t  1.042e-02  1.351e-01   0.077  0.93855    
monthNis:hourly_max_t -1.247e-02  1.320e-01  -0.094  0.92475    
monthOca:hourly_max_t  2.886e-01  1.349e-01   2.140  0.03267 *  
monthŞub:hourly_max_t -3.244e-02  1.320e-01  -0.246  0.80598    
monthTem:hourly_max_t -6.103e-02  1.496e-01  -0.408  0.68333    
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 2.049 on 834 degrees of freedom
Multiple R-squared:  0.5422,	Adjusted R-squared:  0.5257 
F-statistic: 32.92 on 30 and 834 DF,  p-value: < 2.2e-16

	Breusch-Godfrey test for serial correlation of order up to 34

data:  Residuals
LM test = 72.303, df = 34, p-value = 0.0001407


# Plot production for hour 10
ggplot(hour_10_data, aes(x = date, y = production)) +
  geom_line() +
  labs(title = "Hourly Production Data for Hour 10",
       x = "Date",
       y = "Production") +
  theme_minimal()

Warning message in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <c5>”
Warning message in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <9e>”
Warning message in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <c5>”
Warning message in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <9e>”


hour_10_data <- all_data %>%
  filter(hour == 10, date >= start_date & date <= end_date)

forecast_data_hour_10 <- to_be_forecasted %>%
  filter(hour == 10)
forecast_data_hour_10$trend_hour_10 <- nrow(hour_10_data) + 1:nrow(forecast_data_hour_10)

last_production_value <- tail(hour_10_data$production, 1)
forecast_data_hour_10 <- forecast_data_hour_10 %>%
  mutate(lag_1_production = last_production_value)

predictions_hour_10 <- predict(lm_hour_10, newdata = forecast_data_hour_10)
predictions_hour_10[predictions_hour_10 < 0] <- 0

predictions_hour_10_df <- data.frame(
  date = forecast_data_hour_10$date,  
  predicted = predictions_hour_10
)

merged_data <- merge(hour_10_data, predictions_hour_10_df, by = "date")

output_data <- merged_data %>%
  select(date, actual = production, predicted)

output_data <- output_data %>%
  mutate(error = abs(predicted - actual))

wmape <- sum(output_data$error) / sum(output_data$actual) * 100

print(paste("WMAPE:", wmape, "%"))

output_dir <- "C:/Users/ecemozturk/Desktop"
if (!dir.exists(output_dir)) {
  dir.create(output_dir, recursive = TRUE)
}

file_path <- file.path(output_dir, "predictions_hour_10_with_actuals.xlsx")

write_xlsx(output_data, file_path)

print(output_data)

[1] "WMAPE: 17.1336671941216 %"
Key: <date>
           date actual predicted      error
         <IDat>  <num>     <num>      <num>
  1: 2024-02-01   5.20  9.103181 3.90318118
  2: 2024-02-02   9.85  9.791868 0.05813237
  3: 2024-02-03   6.11  7.154666 1.04466562
  4: 2024-02-04   9.83  7.315851 2.51414880
  5: 2024-02-05   6.69  6.172011 0.51798898
 ---                                       
101: 2024-05-11   9.54  8.404196 1.13580386
102: 2024-05-12   0.00  6.531169 6.53116888
103: 2024-05-13   7.85  4.705113 3.14488672
104: 2024-05-14   7.44  7.697921 0.25792078
105: 2024-05-15   9.87  8.350045 1.51995509


# Create a trend variable for hour 11
hour_11_data <- all_data[all_data$hour == 11, ]
hour_11_data <- hour_11_data[,-c(2)]
hour_11_data$trend_hour_11 <- 1:nrow(hour_11_data)

hour_11_data[, lag_19_production := shift(production,19)]
hour_11_data[,lag_19_diff:=production-lag_19_production]
# Remove rows with NA in lagged production to ensure the model can run
hour_11_data <- hour_11_data[!is.na(lag_19_production)]
# Fit linear regression model for hour 11
lm_hour_11 <- lm(production ~+lag_19_production+trend_hour_11+special_period+csnow_surface+dlwrf_surface+tmp_surface+hourly_cloud_average+is.weekend+is.religousday+is.nationalday+month*hourly_max_t, data = hour_11_data)
summary(lm_hour_11)
checkresiduals(lm_hour_11)
plot(lm_hour_11)

Call:
lm(formula = production ~ +lag_19_production + trend_hour_11 + 
    special_period + csnow_surface + dlwrf_surface + tmp_surface + 
    hourly_cloud_average + is.weekend + is.religousday + is.nationalday + 
    month * hourly_max_t, data = hour_11_data)

Residuals:
    Min      1Q  Median      3Q     Max 
-8.7618 -0.6632  0.2996  1.0666  6.7623 

Coefficients: (1 not defined because of singularities)
                        Estimate Std. Error t value Pr(>|t|)    
(Intercept)           -7.716e+01  3.824e+01  -2.018  0.04394 *  
lag_19_production     -8.089e-02  2.593e-02  -3.120  0.00187 ** 
trend_hour_11         -5.759e-04  3.821e-04  -1.507  0.13214    
special_period         3.920e-01  2.550e-01   1.537  0.12472    
csnow_surface         -1.899e+00  5.783e-01  -3.285  0.00106 ** 
dlwrf_surface         -3.286e-02  4.517e-03  -7.276 8.12e-13 ***
tmp_surface            3.146e-01  1.221e-01   2.577  0.01014 *  
hourly_cloud_average  -1.540e-02  5.328e-03  -2.891  0.00394 ** 
is.weekend            -1.661e-01  1.554e-01  -1.069  0.28525    
is.religousday         1.466e-01  4.444e-01   0.330  0.74160    
is.nationalday         2.450e-01  5.779e-01   0.424  0.67173    
monthAra              -7.152e+01  4.482e+01  -1.596  0.11094    
monthEki               3.889e+01  4.401e+01   0.884  0.37711    
monthEyl               3.329e+01  4.175e+01   0.797  0.42550    
monthHaz               3.922e+01  4.427e+01   0.886  0.37591    
monthKas               1.830e+01  4.073e+01   0.449  0.65327    
monthMar               6.715e+01  3.935e+01   1.707  0.08828 .  
monthMay               4.722e+01  4.049e+01   1.166  0.24381    
monthNis               4.275e+01  3.974e+01   1.076  0.28237    
monthOca              -3.682e+01  4.078e+01  -0.903  0.36678    
monthŞub               7.621e+01  3.968e+01   1.921  0.05513 .  
monthTem               4.128e+01  4.461e+01   0.925  0.35500    
hourly_max_t                  NA         NA      NA       NA    
monthAra:hourly_max_t  2.730e-01  1.471e-01   1.856  0.06384 .  
monthEki:hourly_max_t -1.202e-01  1.419e-01  -0.847  0.39706    
monthEyl:hourly_max_t -1.036e-01  1.332e-01  -0.778  0.43658    
monthHaz:hourly_max_t -1.240e-01  1.421e-01  -0.873  0.38299    
monthKas:hourly_max_t -4.877e-02  1.309e-01  -0.372  0.70962    
monthMar:hourly_max_t -2.166e-01  1.259e-01  -1.721  0.08572 .  
monthMay:hourly_max_t -1.520e-01  1.295e-01  -1.174  0.24087    
monthNis:hourly_max_t -1.345e-01  1.269e-01  -1.059  0.28970    
monthOca:hourly_max_t  1.516e-01  1.318e-01   1.150  0.25041    
monthŞub:hourly_max_t -2.490e-01  1.274e-01  -1.955  0.05098 .  
monthTem:hourly_max_t -1.296e-01  1.426e-01  -0.909  0.36342    
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 2.03 on 814 degrees of freedom
Multiple R-squared:  0.5386,	Adjusted R-squared:  0.5204 
F-statistic: 29.69 on 32 and 814 DF,  p-value: < 2.2e-16

	Breusch-Godfrey test for serial correlation of order up to 37

data:  Residuals
LM test = 59.296, df = 37, p-value = 0.01143


# Plot production for hour 11
ggplot(hour_11_data, aes(x = date, y = production)) +
  geom_line() +
  labs(title = "Hourly Production Data for Hour 11",
       x = "Date",
       y = "Production") +
  theme_minimal()

Warning message in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <c5>”
Warning message in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <9e>”
Warning message in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <c5>”
Warning message in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <9e>”


hour_11_data <- all_data %>%
  filter(hour == 11, date >= start_date & date <= end_date)

forecast_data_hour_11 <- to_be_forecasted %>%
  filter(hour == 11)
forecast_data_hour_11$trend_hour_11 <- nrow(hour_11_data) + 1:nrow(forecast_data_hour_11)

last_production_value <- tail(hour_11_data$production, 1)
forecast_data_hour_11 <- forecast_data_hour_11 %>%
  mutate(lag_19_production = last_production_value)

predictions_hour_11 <- predict(lm_hour_11, newdata = forecast_data_hour_11)
predictions_hour_11[predictions_hour_11 < 0] <- 0

predictions_hour_11_df <- data.frame(
  date = forecast_data_hour_11$date,  
  predicted = predictions_hour_11
)

merged_data <- merge(hour_11_data, predictions_hour_11_df, by = "date")

output_data <- merged_data %>%
  select(date, actual = production, predicted)

output_data <- output_data %>%
  mutate(error = abs(predicted - actual))

wmape <- sum(output_data$error) / sum(output_data$actual) * 100

print(paste("WMAPE:", wmape, "%"))

output_dir <- "C:/Users/ecemozturk/Desktop"
if (!dir.exists(output_dir)) {
  dir.create(output_dir, recursive = TRUE)
}

file_path <- file.path(output_dir, "predictions_hour_11_with_actuals.xlsx")

write_xlsx(output_data, file_path)

print(output_data)

[1] "WMAPE: 19.6298861504385 %"
Key: <date>
           date actual predicted     error
         <IDat>  <num>     <num>     <num>
  1: 2024-02-01   3.53  9.113857 5.5838567
  2: 2024-02-02   9.88  9.598790 0.2812103
  3: 2024-02-03   9.09  7.347482 1.7425182
  4: 2024-02-04   6.96  7.403292 0.4432921
  5: 2024-02-05   4.12  5.898651 1.7786511
 ---                                      
101: 2024-05-11   5.76  7.810759 2.0507586
102: 2024-05-12   0.00  5.750518 5.7505177
103: 2024-05-13   8.36  4.728926 3.6310744
104: 2024-05-14   5.84  7.389215 1.5492146
105: 2024-05-15   9.82  8.429268 1.3907323


# Create a trend variable for hour 12
hour_12_data <- all_data[all_data$hour == 12, ]
hour_12_data <- hour_12_data[,-c(2)]
hour_12_data[, lag_1_production := shift(production,1)]
hour_12_data[,lag_1_diff:=production-lag_1_production]
hour_12_data <- hour_12_data[!is.na(lag_1_production)]
hour_12_data$trend_hour_12 <- 1:nrow(hour_12_data)
lm_hour_12 <- lm(production ~ +lag_1_production+trend_hour_12+csnow_surface+dlwrf_surface+tmp_surface+hourly_cloud_average+is.weekend+is.religousday+is.nationalday+month*hourly_max_t , data = hour_12_data)
summary(lm_hour_12)
checkresiduals(lm_hour_12)
plot(lm_hour_12)

Call:
lm(formula = production ~ +lag_1_production + trend_hour_12 + 
    csnow_surface + dlwrf_surface + tmp_surface + hourly_cloud_average + 
    is.weekend + is.religousday + is.nationalday + month * hourly_max_t, 
    data = hour_12_data)

Residuals:
    Min      1Q  Median      3Q     Max 
-8.5394 -0.7375  0.2502  1.1136  6.5160 

Coefficients: (1 not defined because of singularities)
                        Estimate Std. Error t value Pr(>|t|)    
(Intercept)           -4.327e+01  3.850e+01  -1.124  0.26137    
lag_1_production       9.071e-02  2.827e-02   3.209  0.00138 ** 
trend_hour_12         -7.090e-04  3.195e-04  -2.219  0.02677 *  
csnow_surface         -1.375e+00  5.384e-01  -2.554  0.01081 *  
dlwrf_surface         -3.082e-02  4.333e-03  -7.113 2.44e-12 ***
tmp_surface            2.000e-01  1.218e-01   1.641  0.10110    
hourly_cloud_average  -1.622e-02  5.297e-03  -3.062  0.00227 ** 
is.weekend            -3.110e-01  1.531e-01  -2.031  0.04257 *  
is.religousday         1.213e-01  4.404e-01   0.275  0.78305    
is.nationalday         3.109e-01  5.763e-01   0.540  0.58969    
monthAra              -4.290e+01  4.372e+01  -0.981  0.32671    
monthEki               5.402e+00  4.385e+01   0.123  0.90198    
monthEyl               1.075e+01  4.171e+01   0.258  0.79661    
monthHaz               1.580e+01  4.355e+01   0.363  0.71694    
monthKas              -2.455e+00  4.062e+01  -0.060  0.95183    
monthMar               3.075e+01  3.939e+01   0.781  0.43513    
monthMay               3.139e+01  4.056e+01   0.774  0.43920    
monthNis               1.588e+01  3.993e+01   0.398  0.69096    
monthOca              -4.011e+01  4.033e+01  -0.995  0.32026    
monthŞub               4.898e+01  3.964e+01   1.235  0.21701    
monthTem               3.862e+01  4.401e+01   0.878  0.38042    
hourly_max_t                  NA         NA      NA       NA    
monthAra:hourly_max_t  1.574e-01  1.416e-01   1.112  0.26663    
monthEki:hourly_max_t -1.524e-02  1.401e-01  -0.109  0.91336    
monthEyl:hourly_max_t -3.368e-02  1.319e-01  -0.255  0.79852    
monthHaz:hourly_max_t -5.101e-02  1.386e-01  -0.368  0.71283    
monthKas:hourly_max_t  1.198e-02  1.293e-01   0.093  0.92622    
monthMar:hourly_max_t -9.976e-02  1.248e-01  -0.799  0.42442    
monthMay:hourly_max_t -1.038e-01  1.286e-01  -0.807  0.41968    
monthNis:hourly_max_t -5.067e-02  1.265e-01  -0.401  0.68879    
monthOca:hourly_max_t  1.482e-01  1.288e-01   1.151  0.25025    
monthŞub:hourly_max_t -1.658e-01  1.261e-01  -1.316  0.18864    
monthTem:hourly_max_t -1.230e-01  1.394e-01  -0.882  0.37792    
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 2.019 on 833 degrees of freedom
Multiple R-squared:  0.5384,	Adjusted R-squared:  0.5212 
F-statistic: 31.34 on 31 and 833 DF,  p-value: < 2.2e-16

	Breusch-Godfrey test for serial correlation of order up to 36

data:  Residuals
LM test = 32.413, df = 36, p-value = 0.6399


ggplot(hour_12_data, aes(x = date, y = production)) +
  geom_line() +
  labs(title = "Hourly Production Data for Hour 12",
       x = "Date",
       y = "Production") +
  theme_minimal()

Warning message in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <c5>”
Warning message in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <9e>”
Warning message in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <c5>”
Warning message in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <9e>”


hour_12_data <- all_data %>%
  filter(hour == 12, date >= start_date & date <= end_date)

forecast_data_hour_12 <- to_be_forecasted %>%
  filter(hour == 12)
forecast_data_hour_12$trend_hour_12 <- nrow(hour_12_data) + 1:nrow(forecast_data_hour_12)

last_production_value <- tail(hour_12_data$production, 1)
forecast_data_hour_12 <- forecast_data_hour_12 %>%
  mutate(lag_1_production = last_production_value)

predictions_hour_12 <- predict(lm_hour_12, newdata = forecast_data_hour_12)
predictions_hour_12[predictions_hour_12 < 0] <- 0

predictions_hour_12_df <- data.frame(
  date = forecast_data_hour_12$date,  
  predicted = predictions_hour_12
)

merged_data <- merge(hour_12_data, predictions_hour_12_df, by = "date")

output_data <- merged_data %>%
  select(date, actual = production, predicted)

output_data <- output_data %>%
  mutate(error = abs(predicted - actual))

wmape <- sum(output_data$error) / sum(output_data$actual) * 100

print(paste("WMAPE:", wmape, "%"))

output_dir <- "C:/Users/ecemozturk/Desktop"
if (!dir.exists(output_dir)) {
  dir.create(output_dir, recursive = TRUE)
}

file_path <- file.path(output_dir, "predictions_hour_12_with_actuals.xlsx")

write_xlsx(output_data, file_path)

print(output_data)

[1] "WMAPE: 24.145872036254 %"
Key: <date>
           date actual predicted     error
         <IDat>  <num>     <num>     <num>
  1: 2024-02-01   4.92  9.065805 4.1458054
  2: 2024-02-02   9.85  9.499731 0.3502688
  3: 2024-02-03   8.71  7.153421 1.5565785
  4: 2024-02-04   9.19  7.395700 1.7942998
  5: 2024-02-05   2.85  5.925434 3.0754340
 ---                                      
101: 2024-05-11   2.25  8.158062 5.9080618
102: 2024-05-12   0.00  5.995392 5.9953918
103: 2024-05-13   5.34  5.967664 0.6276639
104: 2024-05-14   7.78  8.504438 0.7244383
105: 2024-05-15   7.92  8.874813 0.9548128


# Create a trend variable for hour 13
hour_13_data <- all_data[all_data$hour == 13, ]
hour_13_data <- hour_13_data[,-c(2)]

hour_13_data[, lag_1_production := shift(production,1)]
hour_13_data[,lag_1_diff:=production-lag_1_production]
# Remove rows with NA in lagged production to ensure the model can run
hour_13_data <- hour_13_data[!is.na(lag_1_production)]
# Remove rows with missing values in the hourly_max_t column
hour_13_data$trend_hour_13 <- 1:nrow(hour_13_data)
lm_hour_13 <- lm(production ~+lag_1_production +trend_hour_13+dlwrf_surface+tmp_surface+hourly_cloud_average+is.weekend+is.nationalday+is.publicholiday+month * hourly_max_t , data = hour_13_data)
summary(lm_hour_13)
checkresiduals(lm_hour_13)
plot(lm_hour_13)

Call:
lm(formula = production ~ +lag_1_production + trend_hour_13 + 
    dlwrf_surface + tmp_surface + hourly_cloud_average + is.weekend + 
    is.nationalday + is.publicholiday + month * hourly_max_t, 
    data = hour_13_data)

Residuals:
    Min      1Q  Median      3Q     Max 
-7.7178 -0.9401  0.3124  1.2178  6.2601 

Coefficients: (1 not defined because of singularities)
                        Estimate Std. Error t value Pr(>|t|)    
(Intercept)           -8.568e+01  3.993e+01  -2.146 0.032188 *  
lag_1_production       9.508e-02  2.783e-02   3.416 0.000666 ***
trend_hour_13         -6.315e-04  3.237e-04  -1.951 0.051391 .  
dlwrf_surface         -3.221e-02  4.352e-03  -7.401 3.29e-13 ***
tmp_surface            3.333e-01  1.259e-01   2.647 0.008275 ** 
hourly_cloud_average  -2.541e-02  5.556e-03  -4.573 5.54e-06 ***
is.weekend             3.282e+00  1.184e+00   2.773 0.005670 ** 
is.nationalday         1.910e+00  8.175e-01   2.336 0.019730 *  
is.publicholiday      -3.571e+00  1.194e+00  -2.991 0.002866 ** 
monthAra               1.945e+01  4.467e+01   0.435 0.663442    
monthEki               7.421e+01  4.535e+01   1.636 0.102127    
monthEyl               6.383e+01  4.335e+01   1.472 0.141287    
monthHaz               6.102e+01  4.485e+01   1.360 0.174059    
monthKas               3.433e+01  4.208e+01   0.816 0.414872    
monthMar               6.470e+01  4.067e+01   1.591 0.112032    
monthMay               8.486e+01  4.199e+01   2.021 0.043580 *  
monthNis               4.363e+01  4.147e+01   1.052 0.293125    
monthOca               1.388e+01  4.160e+01   0.334 0.738782    
monthŞub               8.158e+01  4.100e+01   1.990 0.046951 *  
monthTem               7.787e+01  4.461e+01   1.746 0.081257 .  
hourly_max_t                  NA         NA      NA       NA    
monthAra:hourly_max_t -4.841e-02  1.438e-01  -0.337 0.736503    
monthEki:hourly_max_t -2.372e-01  1.444e-01  -1.642 0.100914    
monthEyl:hourly_max_t -2.004e-01  1.367e-01  -1.467 0.142840    
monthHaz:hourly_max_t -1.920e-01  1.423e-01  -1.350 0.177469    
monthKas:hourly_max_t -1.038e-01  1.335e-01  -0.777 0.437101    
monthMar:hourly_max_t -2.034e-01  1.284e-01  -1.584 0.113635    
monthMay:hourly_max_t -2.722e-01  1.328e-01  -2.050 0.040671 *  
monthNis:hourly_max_t -1.350e-01  1.310e-01  -1.031 0.302951    
monthOca:hourly_max_t -2.678e-02  1.323e-01  -0.202 0.839604    
monthŞub:hourly_max_t -2.628e-01  1.298e-01  -2.024 0.043250 *  
monthTem:hourly_max_t -2.451e-01  1.408e-01  -1.741 0.082065 .  
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 2.095 on 834 degrees of freedom
Multiple R-squared:  0.5576,	Adjusted R-squared:  0.5417 
F-statistic: 35.05 on 30 and 834 DF,  p-value: < 2.2e-16

	Breusch-Godfrey test for serial correlation of order up to 35

data:  Residuals
LM test = 56.897, df = 35, p-value = 0.01108


ggplot(hour_13_data, aes(x = date, y = production)) +
  geom_line() +
  labs(title = "Hourly Production Data for Hour 13",
       x = "Date",
       y = "Production") +
  theme_minimal()

Warning message in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <c5>”
Warning message in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <9e>”
Warning message in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <c5>”
Warning message in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <9e>”


hour_13_data <- all_data %>%
  filter(hour == 13, date >= start_date & date <= end_date)

forecast_data_hour_13 <- to_be_forecasted %>%
  filter(hour == 13)
forecast_data_hour_13$trend_hour_13 <- nrow(hour_13_data) + 1:nrow(forecast_data_hour_13)

last_production_value <- tail(hour_13_data$production, 1)
forecast_data_hour_13 <- forecast_data_hour_13 %>%
  mutate(lag_1_production = last_production_value)

predictions_hour_13 <- predict(lm_hour_13, newdata = forecast_data_hour_13)
predictions_hour_13[predictions_hour_13 < 0] <- 0

predictions_hour_13_df <- data.frame(
  date = forecast_data_hour_13$date,  
  predicted = predictions_hour_13
)

merged_data <- merge(hour_13_data, predictions_hour_13_df, by = "date")

output_data <- merged_data %>%
  select(date, actual = production, predicted)

output_data <- output_data %>%
  mutate(error = abs(predicted - actual))

wmape <- sum(output_data$error) / sum(output_data$actual) * 100

print(paste("WMAPE:", wmape, "%"))

output_dir <- "C:/Users/ecemozturk/Desktop"
if (!dir.exists(output_dir)) {
  dir.create(output_dir, recursive = TRUE)
}

file_path <- file.path(output_dir, "predictions_hour_13_with_actuals.xlsx")

write_xlsx(output_data, file_path)

print(output_data)

[1] "WMAPE: 23.9313334909365 %"
Key: <date>
           date actual predicted     error
         <IDat>  <num>     <num>     <num>
  1: 2024-02-01   4.90  8.903164 4.0031635
  2: 2024-02-02   9.85  9.521604 0.3283964
  3: 2024-02-03   8.02  6.644535 1.3754651
  4: 2024-02-04   9.00  7.239842 1.7601575
  5: 2024-02-05   3.01  5.773566 2.7635661
 ---                                      
101: 2024-05-11   3.67  8.253055 4.5830546
102: 2024-05-12   0.00  5.725042 5.7250422
103: 2024-05-13   4.04  5.886979 1.8469792
104: 2024-05-14   9.64  8.951888 0.6881117
105: 2024-05-15   7.18  8.784960 1.6049597


hour_14_data <- all_data[all_data$hour == 14, ]
hour_14_data <- hour_14_data[,-c(2)]
hour_14_data$trend_hour_14 <- 1:nrow(hour_14_data)
hour_14_data[, lag_14_production := shift(production,1)]
hour_14_data[,lag_14_diff:=production-lag_14_production]
# Remove rows with NA in lagged production to ensure the model can run
hour_14_data <- hour_14_data[!is.na(lag_14_production)]
lm_hour_14 <- lm(production ~ +lag_14_production+ special_period+dlwrf_surface+tmp_surface+is.weekend+is.ramadan+is.religousday+is.nationalday+is.publicholiday+hourly_cloud_average+month*hourly_max_t , data = hour_14_data)
summary(lm_hour_14)
checkresiduals(lm_hour_14)
plot(lm_hour_14)

Call:
lm(formula = production ~ +lag_14_production + special_period + 
    dlwrf_surface + tmp_surface + is.weekend + is.ramadan + is.religousday + 
    is.nationalday + is.publicholiday + hourly_cloud_average + 
    month * hourly_max_t, data = hour_14_data)

Residuals:
    Min      1Q  Median      3Q     Max 
-8.4902 -1.0601  0.2299  1.2149  6.5299 

Coefficients: (1 not defined because of singularities)
                        Estimate Std. Error t value Pr(>|t|)    
(Intercept)           -49.229332  33.355821  -1.476  0.14035    
lag_14_production       0.107539   0.028577   3.763  0.00018 ***
special_period          0.407887   0.218537   1.866  0.06233 .  
dlwrf_surface          -0.025869   0.004248  -6.090 1.72e-09 ***
tmp_surface             0.208630   0.105379   1.980  0.04805 *  
is.weekend              3.323860   1.135289   2.928  0.00351 ** 
is.ramadan              0.207241   0.329407   0.629  0.52943    
is.religousday         -0.017660   0.432060  -0.041  0.96741    
is.nationalday          1.884202   0.783760   2.404  0.01643 *  
is.publicholiday       -3.533244   1.145636  -3.084  0.00211 ** 
hourly_cloud_average   -0.030869   0.005373  -5.746 1.28e-08 ***
monthAra               18.715850  38.743513   0.483  0.62917    
monthEki               51.540702  39.514496   1.304  0.19248    
monthEyl               31.333769  37.469047   0.836  0.40325    
monthHaz               -1.556742  39.088677  -0.040  0.96824    
monthKas               24.578372  35.946040   0.684  0.49432    
monthMar               43.898404  34.298011   1.280  0.20093    
monthMay               62.696698  35.725713   1.755  0.07964 .  
monthNis               27.908032  35.149466   0.794  0.42743    
monthOca               -5.164791  35.170800  -0.147  0.88329    
monthŞub               48.080570  34.544043   1.392  0.16434    
monthTem               55.405723  38.165297   1.452  0.14695    
hourly_max_t                  NA         NA      NA       NA    
monthAra:hourly_max_t  -0.062183   0.125821  -0.494  0.62128    
monthEki:hourly_max_t  -0.171335   0.126625  -1.353  0.17640    
monthEyl:hourly_max_t  -0.100101   0.118579  -0.844  0.39882    
monthHaz:hourly_max_t   0.008465   0.124695   0.068  0.94589    
monthKas:hourly_max_t  -0.081753   0.114721  -0.713  0.47628    
monthMar:hourly_max_t  -0.141418   0.108718  -1.301  0.19370    
monthMay:hourly_max_t  -0.204053   0.113489  -1.798  0.07254 .  
monthNis:hourly_max_t  -0.088926   0.111450  -0.798  0.42516    
monthOca:hourly_max_t   0.026587   0.112360   0.237  0.81301    
monthŞub:hourly_max_t  -0.157881   0.109816  -1.438  0.15090    
monthTem:hourly_max_t  -0.175299   0.120809  -1.451  0.14715    
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 2.012 on 832 degrees of freedom
Multiple R-squared:  0.5738,	Adjusted R-squared:  0.5574 
F-statistic:    35 on 32 and 832 DF,  p-value: < 2.2e-16

	Breusch-Godfrey test for serial correlation of order up to 37

data:  Residuals
LM test = 44.803, df = 37, p-value = 0.1771


ggplot(hour_14_data, aes(x = date, y = production)) +
  geom_line() +
  labs(title = "Hourly Production Data for Hour 14",
       x = "Date",
       y = "Production") +
  theme_minimal()

Warning message in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <c5>”
Warning message in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <9e>”
Warning message in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <c5>”
Warning message in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <9e>”


hour_14_data <- all_data %>%
  filter(hour == 14, date >= start_date & date <= end_date)

forecast_data_hour_14 <- to_be_forecasted %>%
  filter(hour == 14)
forecast_data_hour_14$trend_hour_14 <- nrow(hour_14_data) + 1:nrow(forecast_data_hour_14)

last_production_value <- tail(hour_14_data$production, 1)
forecast_data_hour_14 <- forecast_data_hour_14 %>%
  mutate(lag_14_production = last_production_value)

predictions_hour_14 <- predict(lm_hour_14, newdata = forecast_data_hour_14)
predictions_hour_14[predictions_hour_14 < 0] <- 0

predictions_hour_14_df <- data.frame(
  date = forecast_data_hour_14$date,  
  predicted = predictions_hour_14
)

merged_data <- merge(hour_14_data, predictions_hour_14_df, by = "date")

output_data <- merged_data %>%
  select(date, actual = production, predicted)

output_data <- output_data %>%
  mutate(error = abs(predicted - actual))

wmape <- sum(output_data$error) / sum(output_data$actual) * 100

print(paste("WMAPE:", wmape, "%"))

output_dir <- "C:/Users/ecemozturk/Desktop"
if (!dir.exists(output_dir)) {
  dir.create(output_dir, recursive = TRUE)
}

file_path <- file.path(output_dir, "predictions_hour_14_with_actuals.xlsx")

write_xlsx(output_data, file_path)

print(output_data)

[1] "WMAPE: 25.2414535555969 %"
Key: <date>
           date actual predicted     error
         <IDat>  <num>     <num>     <num>
  1: 2024-02-01   5.78  7.594635 1.8146355
  2: 2024-02-02   8.85  8.247988 0.6020120
  3: 2024-02-03   4.37  5.269883 0.8998832
  4: 2024-02-04   3.25  5.966098 2.7160984
  5: 2024-02-05   1.78  4.643374 2.8633739
 ---                                      
101: 2024-05-11   1.91  7.444803 5.5348028
102: 2024-05-12   0.00  5.197135 5.1971350
103: 2024-05-13   5.89  5.293584 0.5964157
104: 2024-05-14   9.70  8.215896 1.4841037
105: 2024-05-15   6.00  7.685520 1.6855205


hour_15_data <- all_data[all_data$hour == 15, ]
hour_15_data <- hour_15_data[,-c(2)]
hour_15_data$trend_hour_15 <- 1:nrow(hour_15_data)
hour_15_data[, lag_15_production := shift(production,1)]
hour_15_data[,lag_15_diff:=production-lag_15_production]
# Remove rows with NA in lagged production to ensure the model can run
hour_15_data <- hour_15_data[!is.na(lag_15_production)]
lm_hour_15 <- lm(production ~+lag_15_production+trend_hour_15+special_period+dswrf_surface+uswrf_top_of_atmosphere+dlwrf_surface+hourly_cloud_average+tmp_surface+is.weekend+is.ramadan+is.publicholiday +month*hourly_max_t , data = hour_15_data)
summary(lm_hour_15)
checkresiduals(lm_hour_15)
plot(lm_hour_15)

Call:
lm(formula = production ~ +lag_15_production + trend_hour_15 + 
    special_period + dswrf_surface + uswrf_top_of_atmosphere + 
    dlwrf_surface + hourly_cloud_average + tmp_surface + is.weekend + 
    is.ramadan + is.publicholiday + month * hourly_max_t, data = hour_15_data)

Residuals:
    Min      1Q  Median      3Q     Max 
-6.2016 -0.8674  0.0785  1.0296  6.1278 

Coefficients: (1 not defined because of singularities)
                          Estimate Std. Error t value Pr(>|t|)    
(Intercept)             -3.027e+01  2.290e+01  -1.322 0.186554    
lag_15_production        1.126e-01  2.954e-02   3.812 0.000148 ***
trend_hour_15           -9.099e-04  3.005e-04  -3.028 0.002542 ** 
special_period           4.393e-01  2.064e-01   2.128 0.033592 *  
dswrf_surface            2.794e-03  1.325e-03   2.108 0.035292 *  
uswrf_top_of_atmosphere  9.910e-03  1.994e-03   4.970 8.15e-07 ***
dlwrf_surface           -2.490e-02  5.054e-03  -4.926 1.01e-06 ***
hourly_cloud_average    -2.727e-02  4.696e-03  -5.806 9.08e-09 ***
tmp_surface              1.325e-01  7.381e-02   1.795 0.073075 .  
is.weekend               1.330e+00  6.940e-01   1.916 0.055651 .  
is.ramadan               4.968e-01  2.717e-01   1.828 0.067875 .  
is.publicholiday        -1.619e+00  6.894e-01  -2.348 0.019120 *  
monthAra                 5.860e+00  2.896e+01   0.202 0.839681    
monthEki                 7.430e+00  2.907e+01   0.256 0.798293    
monthEyl                -1.439e+01  2.749e+01  -0.524 0.600686    
monthHaz                -5.492e+01  2.830e+01  -1.940 0.052663 .  
monthKas                 1.529e+01  2.573e+01   0.594 0.552418    
monthMar                -2.894e+00  2.435e+01  -0.119 0.905416    
monthMay                -9.822e+00  2.627e+01  -0.374 0.708613    
monthNis                -9.130e+00  2.481e+01  -0.368 0.712980    
monthOca                -4.429e+01  2.542e+01  -1.743 0.081788 .  
monthŞub                -1.830e+01  2.448e+01  -0.747 0.455068    
monthTem                -5.457e+00  2.763e+01  -0.197 0.843492    
hourly_max_t                    NA         NA      NA       NA    
monthAra:hourly_max_t   -2.336e-02  9.604e-02  -0.243 0.807928    
monthEki:hourly_max_t   -2.945e-02  9.430e-02  -0.312 0.754848    
monthEyl:hourly_max_t    4.475e-02  8.781e-02   0.510 0.610475    
monthHaz:hourly_max_t    1.811e-01  9.127e-02   1.984 0.047590 *  
monthKas:hourly_max_t   -5.652e-02  8.323e-02  -0.679 0.497289    
monthMar:hourly_max_t    9.090e-03  7.828e-02   0.116 0.907586    
monthMay:hourly_max_t    3.347e-02  8.465e-02   0.395 0.692666    
monthNis:hourly_max_t    2.897e-02  7.948e-02   0.364 0.715603    
monthOca:hourly_max_t    1.582e-01  8.305e-02   1.905 0.057099 .  
monthŞub:hourly_max_t    6.563e-02  7.907e-02   0.830 0.406755    
monthTem:hourly_max_t    1.707e-02  8.818e-02   0.194 0.846550    
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 1.65 on 830 degrees of freedom
  (1 observation deleted due to missingness)
Multiple R-squared:  0.6737,	Adjusted R-squared:  0.6608 
F-statistic: 51.94 on 33 and 830 DF,  p-value: < 2.2e-16

	Breusch-Godfrey test for serial correlation of order up to 38

data:  Residuals
LM test = 50.289, df = 38, p-value = 0.08757


ggplot(hour_15_data, aes(x = date, y = production)) +
  geom_line() +
  labs(title = "Hourly Production Data for Hour 15",
       x = "Date",
       y = "Production") +
  theme_minimal()

Warning message in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <c5>”
Warning message in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <9e>”
Warning message in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <c5>”
Warning message in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <9e>”


# Filter the actual data for hour 15 and the specific date range
hour_15_data <- all_data %>%
  filter(hour == 15, date >= start_date & date <= end_date)

# Filter the to_be_forecasted data for hour 15
forecast_data_hour_15 <- to_be_forecasted %>%
  filter(hour == 15)

# Add the trend variable for hour 15
forecast_data_hour_15$trend_hour_15 <- nrow(hour_15_data) + 1:nrow(forecast_data_hour_15)

# Add the last known production value from hour_15_data to forecast_data_hour_15 for creating the lagged variable
last_production_value <- tail(hour_15_data$production, 1)
forecast_data_hour_15 <- forecast_data_hour_15 %>%
  mutate(lag_15_production = last_production_value)

# Make predictions
predictions_hour_15 <- predict(lm_hour_15, newdata = forecast_data_hour_15)
predictions_hour_15[predictions_hour_15 < 0] <- 0

# Create a dataframe with predictions and corresponding dates
predictions_hour_15_df <- data.frame(
  date = forecast_data_hour_15$date,  # Assuming 'date' column exists in forecast_data_hour_15
  predicted = predictions_hour_15
)

# Merge with actual values
merged_data <- merge(hour_15_data, predictions_hour_15_df, by = "date")

# Select and rename the columns to keep
output_data <- merged_data %>%
  select(date, actual = production, predicted)

# Calculate WMAPE
output_data <- output_data %>%
  mutate(error = abs(predicted - actual))

wmape <- sum(output_data$error) / sum(output_data$actual) * 100

# Print WMAPE
print(paste("WMAPE:", wmape, "%"))

# Ensure the directory exists
output_dir <- "C:/Users/ecemozturk/Desktop"
if (!dir.exists(output_dir)) {
  dir.create(output_dir, recursive = TRUE)
}

# Specify the file path
file_path <- file.path(output_dir, "predictions_hour_15_with_actuals.xlsx")

# Write the combined data to an Excel file
write_xlsx(output_data, file_path)

# Print the combined data
print(output_data)

[1] "WMAPE: 31.6315801980676 %"
Key: <date>
           date actual predicted     error
         <IDat>  <num>     <num>     <num>
  1: 2024-02-01   4.78  4.321723 0.4582772
  2: 2024-02-02   5.64  4.932761 0.7072392
  3: 2024-02-03   3.24  2.428159 0.8118405
  4: 2024-02-04   3.71  2.997038 0.7129616
  5: 2024-02-05   1.88  2.282787 0.4027869
 ---                                      
101: 2024-05-11   1.24  5.956781 4.7167814
102: 2024-05-12   0.00  4.214975 4.2149754
103: 2024-05-13   5.32  4.455304 0.8646957
104: 2024-05-14   7.82  6.907217 0.9127826
105: 2024-05-15   3.11  6.406795 3.2967952


hour_16_data <- all_data[all_data$hour == 16, ]
hour_16_data <- hour_16_data[,-c(2)]
hour_16_data$trend_hour_16 <- 1:nrow(hour_16_data)
hour_16_data[, lag_16_production := shift(production,1)]
hour_16_data[,lag_16_diff:=production-lag_16_production]
# Remove rows with NA in lagged production to ensure the model can run
hour_16_data <- hour_16_data[!is.na(lag_16_production)]
lm_hour_16 <- lm(production ~+lag_16_production+trend_hour_16+ special_period+dswrf_surface+uswrf_top_of_atmosphere+hourly_cloud_average+is.ramadan+month*hourly_max_t, data = hour_16_data)
summary(lm_hour_16)
checkresiduals(lm_hour_16)
plot(lm_hour_16)

Call:
lm(formula = production ~ +lag_16_production + trend_hour_16 + 
    special_period + dswrf_surface + uswrf_top_of_atmosphere + 
    hourly_cloud_average + is.ramadan + month * hourly_max_t, 
    data = hour_16_data)

Residuals:
    Min      1Q  Median      3Q     Max 
-6.2282 -0.6438 -0.0160  0.5996  8.0625 

Coefficients:
                          Estimate Std. Error t value Pr(>|t|)    
(Intercept)              2.935e+00  1.834e+01   0.160 0.872886    
lag_16_production        2.737e-01  3.176e-02   8.616  < 2e-16 ***
trend_hour_16           -8.723e-04  2.400e-04  -3.635 0.000295 ***
special_period           5.022e-01  1.667e-01   3.012 0.002674 ** 
dswrf_surface            5.930e-03  1.040e-03   5.704 1.62e-08 ***
uswrf_top_of_atmosphere  4.940e-03  1.611e-03   3.066 0.002240 ** 
hourly_cloud_average    -7.908e-03  3.141e-03  -2.517 0.012012 *  
is.ramadan               5.014e-01  2.192e-01   2.287 0.022432 *  
monthAra                -1.212e+01  2.413e+01  -0.502 0.615601    
monthEki                 1.546e+01  2.343e+01   0.660 0.509576    
monthEyl                -1.404e+01  2.254e+01  -0.623 0.533494    
monthHaz                -3.036e+01  2.231e+01  -1.361 0.173812    
monthKas                 1.182e+01  2.110e+01   0.560 0.575421    
monthMar                -1.229e+01  1.938e+01  -0.634 0.526199    
monthMay                -2.709e+00  2.052e+01  -0.132 0.894991    
monthNis                -1.475e+01  1.961e+01  -0.752 0.452165    
monthOca                -3.380e+01  2.039e+01  -1.658 0.097778 .  
monthŞub                -2.585e+01  1.950e+01  -1.325 0.185439    
monthTem                -2.035e+01  2.256e+01  -0.902 0.367252    
hourly_max_t            -1.297e-02  5.896e-02  -0.220 0.825912    
monthAra:hourly_max_t    4.147e-02  8.127e-02   0.510 0.609986    
monthEki:hourly_max_t   -5.615e-02  7.697e-02  -0.730 0.465899    
monthEyl:hourly_max_t    4.373e-02  7.291e-02   0.600 0.548823    
monthHaz:hourly_max_t    9.828e-02  7.265e-02   1.353 0.176480    
monthKas:hourly_max_t   -4.340e-02  6.926e-02  -0.627 0.531106    
monthMar:hourly_max_t    3.834e-02  6.301e-02   0.608 0.543040    
monthMay:hourly_max_t    7.239e-03  6.681e-02   0.108 0.913744    
monthNis:hourly_max_t    4.693e-02  6.348e-02   0.739 0.459948    
monthOca:hourly_max_t    1.189e-01  6.741e-02   1.764 0.078062 .  
monthŞub:hourly_max_t    8.877e-02  6.368e-02   1.394 0.163703    
monthTem:hourly_max_t    6.380e-02  7.276e-02   0.877 0.380820    
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 1.33 on 834 degrees of freedom
Multiple R-squared:  0.6828,	Adjusted R-squared:  0.6714 
F-statistic: 59.84 on 30 and 834 DF,  p-value: < 2.2e-16

	Breusch-Godfrey test for serial correlation of order up to 34

data:  Residuals
LM test = 76.913, df = 34, p-value = 3.623e-05


ggplot(hour_16_data, aes(x = date, y = production)) +
  geom_line() +
  labs(title = "Hourly Production Data for Hour 16",
       x = "Date",
       y = "Production") +
  theme_minimal()

Warning message in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <c5>”
Warning message in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <9e>”
Warning message in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <c5>”
Warning message in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <9e>”


# Filter the actual data for hour 16 and the specific date range
hour_16_data <- all_data %>%
  filter(hour == 16, date >= start_date & date <= end_date)

# Filter the to_be_forecasted data for hour 16
forecast_data_hour_16 <- to_be_forecasted %>%
  filter(hour == 16)

# Add the trend variable for hour 16
forecast_data_hour_16$trend_hour_16 <- nrow(hour_16_data) + 1:nrow(forecast_data_hour_16)

# Add the last known production value from hour_16_data to forecast_data_hour_16 for creating the lagged variable
last_production_value <- tail(hour_16_data$production, 1)
forecast_data_hour_16 <- forecast_data_hour_16 %>%
  mutate(lag_16_production = last_production_value)

# Make predictions
predictions_hour_16 <- predict(lm_hour_16, newdata = forecast_data_hour_16)
predictions_hour_16[predictions_hour_16 < 0] <- 0

# Create a dataframe with predictions and corresponding dates
predictions_hour_16_df <- data.frame(
  date = forecast_data_hour_16$date,  # Assuming 'date' column exists in forecast_data_hour_16
  predicted = predictions_hour_16
)

# Merge with actual values
merged_data <- merge(hour_16_data, predictions_hour_16_df, by = "date")

# Select and rename the columns to keep
output_data <- merged_data %>%
  select(date, actual = production, predicted)

# Calculate WMAPE
output_data <- output_data %>%
  mutate(error = abs(predicted - actual))

wmape <- sum(output_data$error) / sum(output_data$actual) * 100

# Print WMAPE
print(paste("WMAPE:", wmape, "%"))

# Ensure the directory exists
output_dir <- "C:/Users/ecemozturk/Desktop"
if (!dir.exists(output_dir)) {
  dir.create(output_dir, recursive = TRUE)
}

# Specify the file path
file_path <- file.path(output_dir, "predictions_hour_16_with_actuals.xlsx")

# Write the combined data to an Excel file
write_xlsx(output_data, file_path)

# Print the combined data
print(output_data)

[1] "WMAPE: 39.6096503905389 %"
Key: <date>
           date actual predicted     error
         <IDat>  <num>     <num>     <num>
  1: 2024-02-01   0.53 0.9404273 0.4104273
  2: 2024-02-02   1.53 1.1416612 0.3883388
  3: 2024-02-03   0.94 0.3487044 0.5912956
  4: 2024-02-04   0.66 0.5928544 0.0671456
  5: 2024-02-05   0.00 0.4192077 0.4192077
 ---                                      
101: 2024-05-11   0.78 2.9014597 2.1214597
102: 2024-05-12   0.00 2.3909400 2.3909400
103: 2024-05-13   3.62 2.3220644 1.2979356
104: 2024-05-14   4.55 3.7036654 0.8463346
105: 2024-05-15   0.00 3.0294696 3.0294696


hour_17_data <- all_data[all_data$hour == 17, ]
hour_17_data <- hour_17_data[,-c(2)]
hour_17_data$trend_hour_17 <- 1:nrow(hour_17_data)
hour_17_data$nm <- as.numeric(format(hour_17_data$date, "%m"))
# Assuming your date column is named 'date_column'
hour_17_data$is_not_winter <- as.numeric(!hour_17_data$nm %in% c(12, 1, 2))
hour_17_data <- as.data.table(hour_17_data)
#hour_17_data <- hour_17_data[production > 0]
#hour_17_data[, log_production := log(production)]
hour_17_data[, lag_17_production := shift(production,1)]
hour_17_data[,lag_17_diff:=production-lag_17_production]
# Remove rows with NA in lagged production to ensure the model can run
hour_17_data <- hour_17_data[!is.na(lag_17_production)]
hour_17_data <- hour_17_data[!is.na(lag_17_diff)]
lm_hour_17 <- lm(production ~+lag_17_production+trend_hour_17+special_period+is.ramadan+is.religousday+uswrf_surface+uswrf_top_of_atmosphere+dswrf_surface+month*hourly_max_t, data = hour_17_data)
summary(lm_hour_17)
checkresiduals(lm_hour_17)
plot(lm_hour_17)

Call:
lm(formula = production ~ +lag_17_production + trend_hour_17 + 
    special_period + is.ramadan + is.religousday + uswrf_surface + 
    uswrf_top_of_atmosphere + dswrf_surface + month * hourly_max_t, 
    data = hour_17_data)

Residuals:
    Min      1Q  Median      3Q     Max 
-3.2207 -0.2810 -0.0078  0.1988  4.0676 

Coefficients:
                          Estimate Std. Error t value Pr(>|t|)    
(Intercept)             -2.0813716 11.2734156  -0.185   0.8536    
lag_17_production        0.4986512  0.0302889  16.463  < 2e-16 ***
trend_hour_17           -0.0007052  0.0001477  -4.774 2.13e-06 ***
special_period           0.2365340  0.0964244   2.453   0.0144 *  
is.ramadan               0.2787440  0.1268847   2.197   0.0283 *  
is.religousday          -0.1523958  0.1673489  -0.911   0.3627    
uswrf_surface           -0.0044092  0.0026311  -1.676   0.0941 .  
uswrf_top_of_atmosphere  0.0012367  0.0011655   1.061   0.2890    
dswrf_surface            0.0030840  0.0012708   2.427   0.0154 *  
monthAra                -0.2137688 14.5499366  -0.015   0.9883    
monthEki                 2.5256493 14.4314141   0.175   0.8611    
monthEyl                 8.8020038 13.7692349   0.639   0.5228    
monthHaz                11.8663195 13.8225292   0.858   0.3909    
monthKas                 6.1699317 13.0493808   0.473   0.6365    
monthMar                 7.6478407 12.1969094   0.627   0.5308    
monthMay                11.6937865 12.8830519   0.908   0.3643    
monthNis                -0.2445660 12.1437403  -0.020   0.9839    
monthOca                -1.0584106 12.3161696  -0.086   0.9315    
monthŞub                 5.8132312 12.3323010   0.471   0.6375    
monthTem                -8.7422014 14.4185303  -0.606   0.5445    
hourly_max_t             0.0062948  0.0369846   0.170   0.8649    
monthAra:hourly_max_t    0.0017844  0.0495986   0.036   0.9713    
monthEki:hourly_max_t   -0.0097040  0.0480400  -0.202   0.8400    
monthEyl:hourly_max_t   -0.0308600  0.0451584  -0.683   0.4946    
monthHaz:hourly_max_t   -0.0409155  0.0454998  -0.899   0.3688    
monthKas:hourly_max_t   -0.0215024  0.0435056  -0.494   0.6213    
monthMar:hourly_max_t   -0.0283603  0.0403228  -0.703   0.4820    
monthMay:hourly_max_t   -0.0400826  0.0424810  -0.944   0.3457    
monthNis:hourly_max_t   -0.0001357  0.0398308  -0.003   0.9973    
monthOca:hourly_max_t    0.0042546  0.0410731   0.104   0.9175    
monthŞub:hourly_max_t   -0.0211684  0.0409375  -0.517   0.6052    
monthTem:hourly_max_t    0.0273820  0.0470553   0.582   0.5608    
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 0.7635 on 833 degrees of freedom
Multiple R-squared:  0.6759,	Adjusted R-squared:  0.6638 
F-statistic: 56.03 on 31 and 833 DF,  p-value: < 2.2e-16

	Breusch-Godfrey test for serial correlation of order up to 35

data:  Residuals
LM test = 147.01, df = 35, p-value = 1.113e-15


ggplot(hour_17_data, aes(x = date, y = production)) +
  geom_line() +
  labs(title = "Hourly Production Data for Hour 17",
       x = "Date",
       y = "Production") +
  theme_minimal()

Warning message in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <c5>”
Warning message in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <9e>”
Warning message in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <c5>”
Warning message in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <9e>”


# Filter the actual data for hour 17 and the specific date range
hour_17_data <- all_data %>%
  filter(hour == 17, date >= start_date & date <= end_date)

# Filter the to_be_forecasted data for hour 17
forecast_data_hour_17 <- to_be_forecasted %>%
  filter(hour == 17)

# Add the trend variable for hour 17
forecast_data_hour_17$trend_hour_17 <- nrow(hour_17_data) + 1:nrow(forecast_data_hour_17)

# Add the last known production value from hour_17_data to forecast_data_hour_17 for creating the lagged variable
last_production_value <- tail(hour_17_data$production, 1)
forecast_data_hour_17 <- forecast_data_hour_17 %>%
  mutate(lag_17_production = last_production_value)

# Make predictions
predictions_hour_17 <- predict(lm_hour_17, newdata = forecast_data_hour_17)
predictions_hour_17[predictions_hour_17 < 0] <- 0

# Create a dataframe with predictions and corresponding dates
predictions_hour_17_df <- data.frame(
  date = forecast_data_hour_17$date,  # Assuming 'date' column exists in forecast_data_hour_17
  predicted = predictions_hour_17
)

# Merge with actual values
merged_data <- merge(hour_17_data, predictions_hour_17_df, by = "date")

# Select and rename the columns to keep
output_data <- merged_data %>%
  select(date, actual = production, predicted)

# Calculate WMAPE
output_data <- output_data %>%
  mutate(error = abs(predicted - actual))

wmape <- sum(output_data$error) / sum(output_data$actual) * 100

# Print WMAPE
print(paste("WMAPE:", wmape, "%"))

# Ensure the directory exists
output_dir <- "C:/Users/ecemozturk/Desktop"
if (!dir.exists(output_dir)) {
  dir.create(output_dir, recursive = TRUE)
}

# Specify the file path
file_path <- file.path(output_dir, "predictions_hour_17_with_actuals.xlsx")

# Write the combined data to an Excel file
write_xlsx(output_data, file_path)

# Print the combined data
print(output_data)

[1] "WMAPE: 61.689983094304 %"
Key: <date>
           date actual predicted     error
         <IDat>  <num>     <num>     <num>
  1: 2024-02-01   0.00 0.3053737 0.3053737
  2: 2024-02-02   0.00 0.3176936 0.3176936
  3: 2024-02-03   0.00 0.2818699 0.2818699
  4: 2024-02-04   0.00 0.2827902 0.2827902
  5: 2024-02-05   0.00 0.1942501 0.1942501
 ---                                      
101: 2024-05-11   0.30 1.0700304 0.7700304
102: 2024-05-12   0.00 1.0176547 1.0176547
103: 2024-05-13   1.30 1.0150746 0.2849254
104: 2024-05-14   1.33 1.3800042 0.0500042
105: 2024-05-15   0.00 1.1531298 1.1531298


hour_18_data <- all_data[all_data$hour == 18, ]
hour_18_data <- hour_18_data[,-c(2)]
# Assuming your dataframe is named 'data' and the production column is named 'production'
#data_18_filtered <- hour_18_data[hour_18_data$production != 0, ]
hour_18_data$trend_hour_18 <- 1:nrow(hour_18_data)
hour_18_data[, lag_18_production := shift(production,1)]
hour_18_data[,lag_18_diff:=production-lag_18_production]
# Remove rows with NA in lagged production to ensure the model can run
hour_18_data <- hour_18_data[!is.na(lag_18_production)]
lm_hour_18 <- lm(production ~+lag_18_production +trend_hour_18+special_period+tmp_surface+dswrf_surface+month*hourly_max_t  , data = hour_18_data)
summary(lm_hour_18)
checkresiduals(lm_hour_18)
plot(lm_hour_18)

Call:
lm(formula = production ~ +lag_18_production + trend_hour_18 + 
    special_period + tmp_surface + dswrf_surface + month * hourly_max_t, 
    data = hour_18_data)

Residuals:
    Min      1Q  Median      3Q     Max 
-0.6744 -0.0467 -0.0040  0.0102  3.7526 

Coefficients: (1 not defined because of singularities)
                        Estimate Std. Error t value Pr(>|t|)   
(Intercept)           -1.370e+01  4.342e+00  -3.155  0.00166 **
lag_18_production      8.499e-02  3.446e-02   2.466  0.01385 * 
trend_hour_18          2.837e-05  4.623e-05   0.614  0.53957   
special_period         1.015e-01  3.311e-02   3.064  0.00226 **
tmp_surface            4.579e-02  1.437e-02   3.186  0.00149 **
dswrf_surface          2.093e-05  1.291e-04   0.162  0.87126   
monthAra               1.374e+01  5.274e+00   2.606  0.00933 **
monthEki               1.297e+01  5.486e+00   2.363  0.01834 * 
monthEyl               1.386e+01  5.222e+00   2.654  0.00811 **
monthHaz               1.621e+01  5.421e+00   2.991  0.00287 **
monthKas               1.366e+01  5.106e+00   2.676  0.00760 **
monthMar               1.392e+01  4.540e+00   3.066  0.00224 **
monthMay               1.224e+01  5.012e+00   2.442  0.01481 * 
monthNis               1.345e+01  4.690e+00   2.869  0.00422 **
monthOca               1.389e+01  4.602e+00   3.018  0.00262 **
monthŞub               1.385e+01  4.524e+00   3.062  0.00227 **
monthTem               7.246e+00  5.681e+00   1.275  0.20254   
hourly_max_t                  NA         NA      NA       NA   
monthAra:hourly_max_t -4.600e-02  1.800e-02  -2.555  0.01079 * 
monthEki:hourly_max_t -4.346e-02  1.853e-02  -2.345  0.01924 * 
monthEyl:hourly_max_t -4.656e-02  1.739e-02  -2.677  0.00758 **
monthHaz:hourly_max_t -5.379e-02  1.810e-02  -2.972  0.00304 **
monthKas:hourly_max_t -4.581e-02  1.727e-02  -2.653  0.00813 **
monthMar:hourly_max_t -4.664e-02  1.511e-02  -3.086  0.00209 **
monthMay:hourly_max_t -4.056e-02  1.676e-02  -2.420  0.01574 * 
monthNis:hourly_max_t -4.498e-02  1.561e-02  -2.881  0.00407 **
monthOca:hourly_max_t -4.653e-02  1.541e-02  -3.019  0.00261 **
monthŞub:hourly_max_t -4.641e-02  1.509e-02  -3.076  0.00217 **
monthTem:hourly_max_t -2.342e-02  1.884e-02  -1.243  0.21417   
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 0.2659 on 837 degrees of freedom
Multiple R-squared:  0.1977,	Adjusted R-squared:  0.1719 
F-statistic:  7.64 on 27 and 837 DF,  p-value: < 2.2e-16

	Breusch-Godfrey test for serial correlation of order up to 32

data:  Residuals
LM test = 207.6, df = 32, p-value < 2.2e-16


ggplot(hour_18_data, aes(x = date, y = production)) +
  geom_line() +
  labs(title = "Hourly Production Data for Hour 18",
       x = "Date",
       y = "Production") +
  theme_minimal()

Warning message in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <c5>”
Warning message in grid.Call(C_textBounds, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <9e>”
Warning message in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <c5>”
Warning message in grid.Call.graphics(C_text, as.graphicsAnnot(x$label), x$x, x$y, :
“conversion failure on 'Şub' in 'mbcsToSbcs': dot substituted for <9e>”


# Filter the actual data for hour 18 and the specific date range
hour_18_data <- all_data %>%
  filter(hour == 18, date >= start_date & date <= end_date)

# Filter the to_be_forecasted data for hour 18
forecast_data_hour_18 <- to_be_forecasted %>%
  filter(hour == 18)

# Add the trend variable for hour 18
forecast_data_hour_18$trend_hour_18 <- nrow(hour_18_data) + 1:nrow(forecast_data_hour_18)

# Add the last known production value from hour_18_data to forecast_data_hour_18 for creating the lagged variable
last_production_value <- tail(hour_18_data$production, 1)
forecast_data_hour_18 <- forecast_data_hour_18 %>%
  mutate(lag_18_production = last_production_value)

# Make predictions
predictions_hour_18 <- predict(lm_hour_18, newdata = forecast_data_hour_18)
predictions_hour_18[predictions_hour_18 < 0] <- 0

# Create a dataframe with predictions and corresponding dates
predictions_hour_18_df <- data.frame(
  date = forecast_data_hour_18$date,  # Assuming 'date' column exists in forecast_data_hour_18
  predicted = predictions_hour_18
)

# Merge with actual values
merged_data <- merge(hour_18_data, predictions_hour_18_df, by = "date")

# Select and rename the columns to keep
output_data <- merged_data %>%
  select(date, actual = production, predicted)

# Calculate WMAPE
output_data <- output_data %>%
  mutate(error = abs(predicted - actual))

wmape <- sum(output_data$error) / sum(output_data$actual) * 100

# Print WMAPE
print(paste("WMAPE:", wmape, "%"))

# Print the combined data
print(output_data)

[1] "WMAPE: 87.6958599729125 %"
Key: <date>
           date actual  predicted      error
         <IDat>  <num>      <num>      <num>
  1: 2024-02-01   0.00 0.00000000 0.00000000
  2: 2024-02-02   0.00 0.00000000 0.00000000
  3: 2024-02-03   0.00 0.00000000 0.00000000
  4: 2024-02-04   0.00 0.00000000 0.00000000
  5: 2024-02-05   0.00 0.00000000 0.00000000
 ---                                        
101: 2024-05-11   0.00 0.05853033 0.05853033
102: 2024-05-12   0.00 0.04082760 0.04082760
103: 2024-05-13   0.16 0.04495209 0.11504791
104: 2024-05-14   0.14 0.05715345 0.08284655
105: 2024-05-15   0.00 0.06877722 0.06877722

date	hour	lat	lon	dswrf_surface	tcdc_low.cloud.layer	tcdc_middle.cloud.layer	tcdc_high.cloud.layer	tcdc_entire.atmosphere	uswrf_top_of_atmosphere	csnow_surface	dlwrf_surface	uswrf_surface	tmp_surface
<IDate>	<int>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<int>	<dbl>	<dbl>	<dbl>
2022-01-13	23	37.75	35.00	0	56.0	90.1	NA	98.1	0	0	223.900	0	262.257
2022-01-13	23	38.75	35.25	0	100.0	10.0	NA	100.0	0	0	250.200	0	271.157
2022-01-13	23	37.75	34.50	0	36.5	85.5	NA	98.9	0	0	242.400	0	270.657
2022-01-13	23	38.50	35.25	0	29.9	51.2	NA	94.6	0	0	226.500	0	266.857
2022-01-13	23	37.75	35.50	0	46.7	99.8	NA	99.9	0	0	253.200	0	267.357
2022-01-13	23	38.75	35.50	0	100.0	8.7	NA	100.0	0	0	236.000	0	270.857
2022-01-13	23	38.00	35.25	0	99.2	76.1	NA	100.0	0	1	230.200	0	259.557
2022-01-13	23	38.25	35.50	0	99.5	75.0	NA	99.8	0	1	267.300	0	268.257
2022-01-13	23	38.25	35.25	0	100.0	68.7	NA	100.0	0	1	295.200	0	271.657
2022-01-13	23	37.75	35.25	0	62.9	86.2	NA	90.7	0	0	233.200	0	262.957
2022-01-13	23	38.25	35.00	0	99.9	47.3	NA	100.0	0	1	268.200	0	269.357
2022-01-13	23	38.00	35.50	0	87.7	98.1	NA	99.1	0	1	255.500	0	265.857
2022-01-13	23	38.00	35.00	0	99.6	63.7	NA	100.0	0	1	258.400	0	263.857
2022-01-13	23	37.75	34.75	0	10.6	93.3	NA	97.0	0	0	229.800	0	265.857
2022-01-13	23	38.50	34.50	0	100.0	9.7	NA	100.0	0	1	240.100	0	267.657
2022-01-13	23	38.50	34.75	0	98.7	9.3	NA	100.0	0	1	236.900	0	266.957
2022-01-13	23	38.50	35.00	0	99.4	7.0	NA	99.6	0	0	233.100	0	267.057
2022-01-13	23	38.50	35.50	0	98.5	51.7	NA	100.0	0	1	223.600	0	259.357
2022-01-13	23	38.75	34.50	0	98.0	0.0	NA	98.2	0	0	238.700	0	270.457
2022-01-13	23	38.25	34.50	0	100.0	5.0	NA	100.0	0	1	249.100	0	266.257
2022-01-13	23	38.75	35.00	0	95.6	18.7	NA	96.1	0	1	242.400	0	270.357
2022-01-13	23	38.00	34.75	0	96.8	47.7	NA	100.0	0	1	247.400	0	265.657
2022-01-13	23	38.75	34.75	0	85.2	2.2	NA	86.1	0	0	239.300	0	270.657
2022-01-13	23	38.25	34.75	0	100.0	9.8	NA	100.0	0	1	254.900	0	268.657
2022-01-13	23	38.00	34.50	0	6.3	56.0	NA	100.0	0	0	221.700	0	265.157
2022-01-27	3	38.75	34.50	0	NA	NA	39.1	100.0	0	1	230.487	0	257.784
2022-01-27	3	38.25	35.50	0	NA	NA	100.0	100.0	0	1	248.887	0	258.784
2022-01-27	3	38.00	35.50	0	NA	NA	100.0	100.0	0	0	245.987	0	258.984
2022-01-27	3	37.75	34.75	0	NA	NA	100.0	100.0	0	0	224.287	0	257.284
2022-01-27	3	38.50	34.50	0	NA	NA	79.0	100.0	0	1	224.387	0	256.084
⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮
2022-03-23	7	37.75	34.50	0.56	9.6	3.5	0.0	NA	0.800	0	205.727	0.192	268.461
2022-03-23	7	38.75	34.50	0.56	5.7	2.1	0.0	NA	0.736	0	205.127	0.144	269.361
2022-03-23	7	38.25	34.75	0.72	19.1	40.5	0.0	NA	1.216	0	193.527	0.528	258.261
2022-03-23	7	38.50	34.50	0.60	6.6	13.9	0.0	NA	1.024	0	190.827	0.432	256.561
2022-03-23	7	38.00	34.50	0.60	10.3	40.9	0.0	NA	0.896	0	198.327	0.304	261.761
2022-04-04	10	38.75	35.00	544.02	0.0	0.0	97.6	98.3	NA	0	266.397	NA	291.531
2022-04-04	10	37.75	35.50	579.18	0.0	0.0	72.6	80.7	NA	0	269.997	NA	293.231
2022-04-04	10	37.75	35.00	615.56	0.0	0.0	5.0	10.8	NA	0	238.297	NA	283.331
2022-04-04	10	37.75	35.25	614.90	0.0	0.0	20.0	26.5	NA	0	237.597	NA	286.931
2022-04-04	10	38.00	35.25	616.42	0.0	0.0	5.0	10.5	NA	0	232.197	NA	282.231
2022-04-04	10	38.50	35.50	612.84	0.0	0.0	9.1	12.2	NA	0	235.097	NA	277.831
2022-04-04	10	38.25	34.50	590.28	0.0	0.0	10.1	16.6	NA	0	251.497	NA	288.831
2022-04-04	10	38.25	35.00	591.38	0.0	0.0	33.0	37.2	NA	0	260.197	NA	292.231
2022-04-04	10	38.50	35.00	587.14	0.0	0.0	48.6	52.8	NA	0	255.197	NA	289.131
2022-04-04	10	38.25	35.25	588.62	0.0	0.0	6.0	9.9	NA	0	268.997	NA	292.131
2022-04-04	10	38.50	34.50	581.62	0.0	0.0	37.1	41.2	NA	0	257.197	NA	289.731
2022-04-04	10	38.75	35.50	546.76	0.0	0.0	85.2	88.5	NA	0	269.097	NA	291.931
2022-04-04	10	38.00	35.00	603.96	0.0	0.0	5.0	10.8	NA	0	243.897	NA	286.031
2022-04-04	10	38.75	34.75	555.22	0.0	0.0	92.4	93.6	NA	0	264.697	NA	292.331
2022-04-04	10	38.75	34.50	548.94	0.0	0.0	99.5	99.6	NA	0	265.897	NA	292.531
2022-04-04	10	38.25	35.50	592.34	0.0	0.0	5.0	9.2	NA	0	261.197	NA	291.331
2022-04-04	10	37.75	34.75	601.20	0.0	0.0	7.6	12.4	NA	0	254.597	NA	288.431
2022-04-04	10	38.75	35.25	531.68	0.0	0.0	100.0	100.0	NA	0	270.297	NA	291.531
2022-04-04	10	38.25	34.75	589.52	0.0	0.0	46.7	52.7	NA	0	258.097	NA	292.631
2022-04-04	10	38.50	34.75	585.10	0.0	0.0	52.9	57.4	NA	0	253.097	NA	289.231
2022-04-04	10	38.50	35.25	591.58	0.0	0.0	8.7	14.5	NA	0	255.897	NA	290.131
2022-04-04	10	38.00	34.75	596.30	0.0	0.0	5.0	11.1	NA	0	252.797	NA	288.931
2022-04-04	10	38.00	34.50	594.70	0.0	0.0	5.7	12.6	NA	0	251.297	NA	288.131
2022-04-04	10	38.00	35.50	592.68	0.0	0.0	70.4	77.8	NA	0	256.497	NA	290.531
2022-04-04	10	37.75	34.50	587.90	0.0	0.0	17.8	23.8	NA	0	269.797	NA	294.931

date	hour	dswrf_surface	tcdc_low.cloud.layer	tcdc_middle.cloud.layer	tcdc_high.cloud.layer	tcdc_entire.atmosphere	uswrf_top_of_atmosphere	csnow_surface	dlwrf_surface	uswrf_surface	tmp_surface
<IDate>	<int>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>
2022-01-01	4	0.0000	2.384	5.944	4.604	14.296	0.00000	0.00	227.9990	0.00000	269.220
2022-01-01	5	0.0000	2.784	4.324	10.636	19.272	0.00000	0.00	227.7740	0.00000	269.104
2022-01-01	6	0.0000	2.964	5.372	11.688	21.772	0.00000	0.00	227.7640	0.00000	269.035
2022-01-01	7	0.0000	3.284	9.212	20.736	31.992	0.00000	0.00	228.1960	0.00000	269.001
2022-01-01	8	0.0000	3.672	11.252	26.432	38.376	0.00000	0.00	228.6570	0.00000	269.002
2022-01-01	9	7.3688	4.120	10.880	35.088	45.856	8.96704	0.00	229.4160	2.41280	271.634
2022-01-01	10	180.1384	4.180	13.748	80.764	85.364	135.20448	0.00	237.2910	59.18848	275.786
2022-01-01	11	254.7928	3.972	16.520	69.392	75.468	152.98944	0.00	237.3870	82.53120	278.553
2022-01-01	12	312.5520	4.408	23.468	60.516	70.368	167.55392	0.00	238.3870	99.73056	280.215
2022-01-01	13	347.4144	5.348	32.996	54.852	70.452	180.69312	0.00	240.8870	109.43168	280.609
2022-01-01	14	364.6544	6.772	39.460	45.880	71.136	187.03872	0.00	243.0150	113.51040	280.277
2022-01-01	15	362.2984	9.116	43.452	39.864	71.612	188.88640	0.00	245.2960	112.10304	279.165
2022-01-01	16	221.4584	37.100	82.296	12.680	89.116	168.56448	0.00	260.7560	66.06976	277.059
2022-01-01	17	157.6448	37.032	78.252	17.376	85.688	133.18080	0.00	258.9600	47.22432	273.467
2022-01-01	18	107.3080	35.876	74.752	27.016	83.300	92.25216	0.00	257.2560	32.11968	271.659
2022-01-01	19	80.4792	35.080	72.480	37.268	83.508	69.18720	0.00	256.5060	24.08960	271.537
2022-01-01	20	64.3864	36.096	75.236	44.544	85.644	55.34976	0.00	257.7050	19.27104	271.715
2022-01-01	21	53.6528	37.028	77.632	49.756	87.520	46.12352	0.00	259.4850	16.06016	271.855
2022-01-01	22	0.0000	49.732	90.688	68.948	98.276	0.00000	0.00	272.8750	0.00000	272.029
2022-01-01	23	0.0000	56.532	92.580	69.140	98.380	0.00000	0.00	274.7220	0.00000	272.231
2022-01-02	0	0.0000	66.812	94.140	76.844	98.760	0.00000	0.04	280.7740	0.00000	272.513
2022-01-02	1	0.0000	71.096	94.212	80.404	98.492	0.00000	0.20	283.7540	0.00000	272.616
2022-01-02	2	0.0000	74.580	94.000	84.192	98.792	0.00000	0.32	287.3460	0.00000	273.024
2022-01-02	3	0.0000	77.660	94.556	86.644	98.980	0.00000	0.48	291.0600	0.00000	273.426
2022-01-02	4	0.0000	93.816	91.588	99.228	100.000	0.00000	0.60	312.8040	0.00000	273.605
2022-01-02	5	0.0000	92.820	88.048	99.388	100.000	0.00000	0.60	311.7720	0.00000	273.537
2022-01-02	6	0.0000	93.304	89.960	99.588	100.000	0.00000	0.68	312.2060	0.00000	273.710
2022-01-02	7	0.0000	93.492	91.856	99.696	100.000	0.00000	0.64	312.9950	0.00000	273.837
2022-01-02	8	0.0000	93.520	93.408	99.756	100.000	0.00000	0.64	313.6920	0.00000	273.885
2022-01-02	9	1.1968	93.872	94.488	99.656	100.000	13.37856	0.64	314.2307	0.53568	274.263
⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮
2024-05-19	16	741.68400	3.748	12.400	13.308	26.132	184.14656	0	315.8730	129.28704	300.115
2024-05-19	17	653.24720	4.836	13.920	15.876	30.608	177.16928	0	315.2560	119.13024	297.226
2024-05-19	18	551.01840	6.664	17.332	22.384	39.596	171.59360	0	315.9100	104.58816	293.967
2024-05-19	19	451.50800	7.392	17.692	33.540	48.368	157.55328	0	315.3600	87.64096	290.504
2024-05-19	20	365.32160	8.288	16.824	44.928	56.940	132.32256	0	314.6400	70.99328	287.699
2024-05-19	21	304.43328	7.720	14.724	52.296	62.416	110.26688	0	313.1420	59.15904	286.748
2024-05-19	22	0.00000	0.932	0.144	86.084	86.800	0.00000	0	302.1770	0.00000	286.008
2024-05-19	23	0.00000	0.544	0.144	75.572	76.660	0.00000	0	300.4080	0.00000	285.319
2024-05-20	0	0.00000	0.364	0.300	64.076	65.364	0.00000	0	298.4437	0.00000	284.668
2024-05-20	1	0.00000	0.268	0.300	52.524	53.820	0.00000	0	296.7120	0.00000	284.064
2024-05-20	2	0.00000	0.216	0.304	44.528	45.840	0.00000	0	295.4120	0.00000	283.597
2024-05-20	3	0.00000	0.180	0.464	39.212	40.740	0.00000	0	294.4750	0.00000	283.175
2024-05-20	4	0.00000	0.104	3.632	33.512	37.208	0.00000	0	289.2320	0.00000	282.832
2024-05-20	5	0.00000	0.260	3.596	43.000	46.432	0.00000	0	289.8500	0.00000	282.536
2024-05-20	6	3.35600	0.904	5.816	39.872	45.740	4.07168	0	290.3950	0.77888	283.660
2024-05-20	7	35.98848	1.968	7.488	33.316	41.208	23.18208	0	290.5790	9.36448	288.299
2024-05-20	8	88.90048	3.752	11.000	29.284	41.020	48.74688	0	293.7390	20.81152	292.868
2024-05-20	9	148.04160	5.272	15.512	26.252	42.420	77.78240	0	297.8070	31.73120	296.414
2024-05-20	10	623.06000	16.520	34.476	12.796	45.516	239.66880	0	318.9140	109.52576	300.288
2024-05-20	11	697.54480	10.808	32.828	11.096	43.596	243.46960	0	321.0400	118.63680	303.036
2024-05-20	12	757.79520	10.524	30.456	11.920	42.148	243.22688	0	322.7000	125.46112	304.884
2024-05-20	13	786.57280	12.068	29.524	13.816	41.716	251.42784	0	325.2360	128.43776	305.034
2024-05-20	14	780.91120	14.284	31.840	17.044	45.484	270.32896	0	329.1890	127.48992	303.220
2024-05-20	15	748.39680	17.028	35.120	22.488	52.052	294.02496	0	333.3480	123.05600	300.109
2024-05-20	16	428.61520	34.728	57.824	49.888	92.296	423.56096	0	358.2960	79.20320	297.543
2024-05-20	17	367.40000	34.492	62.112	44.068	93.800	396.17088	0	359.0440	69.55136	295.564
2024-05-20	18	312.24160	34.684	66.848	42.728	94.424	354.52992	0	358.4140	60.48000	293.613
2024-05-20	19	261.56960	31.080	66.276	49.852	93.684	303.41760	0	354.7380	51.42720	291.316
2024-05-20	20	212.68736	30.616	64.848	57.588	93.680	249.77856	0	351.5110	41.86048	288.794
2024-05-20	21	177.24096	29.016	62.472	63.936	94.160	208.14784	0	347.9070	34.88192	288.011

date	hour	production	dswrf_surface	tcdc_low.cloud.layer	tcdc_middle.cloud.layer	tcdc_high.cloud.layer	tcdc_entire.atmosphere	uswrf_top_of_atmosphere	csnow_surface	dlwrf_surface	uswrf_surface	tmp_surface
<IDate>	<int>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>
2022-01-01	4	0.00	0.0000	2.384	5.944	4.604	14.296	0.00000	0.00	227.9990	0.00000	269.220
2022-01-01	5	0.00	0.0000	2.784	4.324	10.636	19.272	0.00000	0.00	227.7740	0.00000	269.104
2022-01-01	6	0.00	0.0000	2.964	5.372	11.688	21.772	0.00000	0.00	227.7640	0.00000	269.035
2022-01-01	7	0.00	0.0000	3.284	9.212	20.736	31.992	0.00000	0.00	228.1960	0.00000	269.001
2022-01-01	8	3.40	0.0000	3.672	11.252	26.432	38.376	0.00000	0.00	228.6570	0.00000	269.002
2022-01-01	9	6.80	7.3688	4.120	10.880	35.088	45.856	8.96704	0.00	229.4160	2.41280	271.634
2022-01-01	10	9.38	180.1384	4.180	13.748	80.764	85.364	135.20448	0.00	237.2910	59.18848	275.786
2022-01-01	11	7.65	254.7928	3.972	16.520	69.392	75.468	152.98944	0.00	237.3870	82.53120	278.553
2022-01-01	12	6.80	312.5520	4.408	23.468	60.516	70.368	167.55392	0.00	238.3870	99.73056	280.215
2022-01-01	13	5.10	347.4144	5.348	32.996	54.852	70.452	180.69312	0.00	240.8870	109.43168	280.609
2022-01-01	14	5.10	364.6544	6.772	39.460	45.880	71.136	187.03872	0.00	243.0150	113.51040	280.277
2022-01-01	15	1.70	362.2984	9.116	43.452	39.864	71.612	188.88640	0.00	245.2960	112.10304	279.165
2022-01-01	16	0.00	221.4584	37.100	82.296	12.680	89.116	168.56448	0.00	260.7560	66.06976	277.059
2022-01-01	17	0.00	157.6448	37.032	78.252	17.376	85.688	133.18080	0.00	258.9600	47.22432	273.467
2022-01-01	18	0.00	107.3080	35.876	74.752	27.016	83.300	92.25216	0.00	257.2560	32.11968	271.659
2022-01-01	19	0.00	80.4792	35.080	72.480	37.268	83.508	69.18720	0.00	256.5060	24.08960	271.537
2022-01-01	20	0.00	64.3864	36.096	75.236	44.544	85.644	55.34976	0.00	257.7050	19.27104	271.715
2022-01-01	21	0.00	53.6528	37.028	77.632	49.756	87.520	46.12352	0.00	259.4850	16.06016	271.855
2022-01-01	22	0.00	0.0000	49.732	90.688	68.948	98.276	0.00000	0.00	272.8750	0.00000	272.029
2022-01-01	23	0.00	0.0000	56.532	92.580	69.140	98.380	0.00000	0.00	274.7220	0.00000	272.231
2022-01-02	0	0.00	0.0000	66.812	94.140	76.844	98.760	0.00000	0.04	280.7740	0.00000	272.513
2022-01-02	1	0.00	0.0000	71.096	94.212	80.404	98.492	0.00000	0.20	283.7540	0.00000	272.616
2022-01-02	2	0.00	0.0000	74.580	94.000	84.192	98.792	0.00000	0.32	287.3460	0.00000	273.024
2022-01-02	3	0.00	0.0000	77.660	94.556	86.644	98.980	0.00000	0.48	291.0600	0.00000	273.426
2022-01-02	4	0.00	0.0000	93.816	91.588	99.228	100.000	0.00000	0.60	312.8040	0.00000	273.605
2022-01-02	5	0.00	0.0000	92.820	88.048	99.388	100.000	0.00000	0.60	311.7720	0.00000	273.537
2022-01-02	6	0.00	0.0000	93.304	89.960	99.588	100.000	0.00000	0.68	312.2060	0.00000	273.710
2022-01-02	7	0.00	0.0000	93.492	91.856	99.696	100.000	0.00000	0.64	312.9950	0.00000	273.837
2022-01-02	8	0.00	0.0000	93.520	93.408	99.756	100.000	0.00000	0.64	313.6920	0.00000	273.885
2022-01-02	9	0.85	1.1968	93.872	94.488	99.656	100.000	13.37856	0.64	314.2307	0.53568	274.263
⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮
2024-05-19	16	NA	741.68400	3.748	12.400	13.308	26.132	184.14656	0	315.8730	129.28704	300.115
2024-05-19	17	NA	653.24720	4.836	13.920	15.876	30.608	177.16928	0	315.2560	119.13024	297.226
2024-05-19	18	NA	551.01840	6.664	17.332	22.384	39.596	171.59360	0	315.9100	104.58816	293.967
2024-05-19	19	NA	451.50800	7.392	17.692	33.540	48.368	157.55328	0	315.3600	87.64096	290.504
2024-05-19	20	NA	365.32160	8.288	16.824	44.928	56.940	132.32256	0	314.6400	70.99328	287.699
2024-05-19	21	NA	304.43328	7.720	14.724	52.296	62.416	110.26688	0	313.1420	59.15904	286.748
2024-05-19	22	NA	0.00000	0.932	0.144	86.084	86.800	0.00000	0	302.1770	0.00000	286.008
2024-05-19	23	NA	0.00000	0.544	0.144	75.572	76.660	0.00000	0	300.4080	0.00000	285.319
2024-05-20	0	NA	0.00000	0.364	0.300	64.076	65.364	0.00000	0	298.4437	0.00000	284.668
2024-05-20	1	NA	0.00000	0.268	0.300	52.524	53.820	0.00000	0	296.7120	0.00000	284.064
2024-05-20	2	NA	0.00000	0.216	0.304	44.528	45.840	0.00000	0	295.4120	0.00000	283.597
2024-05-20	3	NA	0.00000	0.180	0.464	39.212	40.740	0.00000	0	294.4750	0.00000	283.175
2024-05-20	4	NA	0.00000	0.104	3.632	33.512	37.208	0.00000	0	289.2320	0.00000	282.832
2024-05-20	5	NA	0.00000	0.260	3.596	43.000	46.432	0.00000	0	289.8500	0.00000	282.536
2024-05-20	6	NA	3.35600	0.904	5.816	39.872	45.740	4.07168	0	290.3950	0.77888	283.660
2024-05-20	7	NA	35.98848	1.968	7.488	33.316	41.208	23.18208	0	290.5790	9.36448	288.299
2024-05-20	8	NA	88.90048	3.752	11.000	29.284	41.020	48.74688	0	293.7390	20.81152	292.868
2024-05-20	9	NA	148.04160	5.272	15.512	26.252	42.420	77.78240	0	297.8070	31.73120	296.414
2024-05-20	10	NA	623.06000	16.520	34.476	12.796	45.516	239.66880	0	318.9140	109.52576	300.288
2024-05-20	11	NA	697.54480	10.808	32.828	11.096	43.596	243.46960	0	321.0400	118.63680	303.036
2024-05-20	12	NA	757.79520	10.524	30.456	11.920	42.148	243.22688	0	322.7000	125.46112	304.884
2024-05-20	13	NA	786.57280	12.068	29.524	13.816	41.716	251.42784	0	325.2360	128.43776	305.034
2024-05-20	14	NA	780.91120	14.284	31.840	17.044	45.484	270.32896	0	329.1890	127.48992	303.220
2024-05-20	15	NA	748.39680	17.028	35.120	22.488	52.052	294.02496	0	333.3480	123.05600	300.109
2024-05-20	16	NA	428.61520	34.728	57.824	49.888	92.296	423.56096	0	358.2960	79.20320	297.543
2024-05-20	17	NA	367.40000	34.492	62.112	44.068	93.800	396.17088	0	359.0440	69.55136	295.564
2024-05-20	18	NA	312.24160	34.684	66.848	42.728	94.424	354.52992	0	358.4140	60.48000	293.613
2024-05-20	19	NA	261.56960	31.080	66.276	49.852	93.684	303.41760	0	354.7380	51.42720	291.316
2024-05-20	20	NA	212.68736	30.616	64.848	57.588	93.680	249.77856	0	351.5110	41.86048	288.794
2024-05-20	21	NA	177.24096	29.016	62.472	63.936	94.160	208.14784	0	347.9070	34.88192	288.011

date	hour	production	dswrf_surface	tcdc_low.cloud.layer	tcdc_middle.cloud.layer	tcdc_high.cloud.layer	tcdc_entire.atmosphere	uswrf_top_of_atmosphere	csnow_surface	dlwrf_surface	uswrf_surface	tmp_surface	hourly_cloud_average	hourly_max_t
<IDate>	<int>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>
2022-01-01	4	0.00	0.0000	2.384	5.944	4.604	14.296	0.00000	0.00	227.9990	0.00000	269.220	6.807	269.220
2022-01-01	5	0.00	0.0000	2.784	4.324	10.636	19.272	0.00000	0.00	227.7740	0.00000	269.104	9.254	269.104
2022-01-01	6	0.00	0.0000	2.964	5.372	11.688	21.772	0.00000	0.00	227.7640	0.00000	269.035	10.449	269.035
2022-01-01	7	0.00	0.0000	3.284	9.212	20.736	31.992	0.00000	0.00	228.1960	0.00000	269.001	16.306	269.001
2022-01-01	8	3.40	0.0000	3.672	11.252	26.432	38.376	0.00000	0.00	228.6570	0.00000	269.002	19.933	269.002
2022-01-01	9	6.80	7.3688	4.120	10.880	35.088	45.856	8.96704	0.00	229.4160	2.41280	271.634	23.986	271.634
2022-01-01	10	9.38	180.1384	4.180	13.748	80.764	85.364	135.20448	0.00	237.2910	59.18848	275.786	46.014	275.786
2022-01-01	11	7.65	254.7928	3.972	16.520	69.392	75.468	152.98944	0.00	237.3870	82.53120	278.553	41.338	278.553
2022-01-01	12	6.80	312.5520	4.408	23.468	60.516	70.368	167.55392	0.00	238.3870	99.73056	280.215	39.690	280.215
2022-01-01	13	5.10	347.4144	5.348	32.996	54.852	70.452	180.69312	0.00	240.8870	109.43168	280.609	40.912	280.609
2022-01-01	14	5.10	364.6544	6.772	39.460	45.880	71.136	187.03872	0.00	243.0150	113.51040	280.277	40.812	280.277
2022-01-01	15	1.70	362.2984	9.116	43.452	39.864	71.612	188.88640	0.00	245.2960	112.10304	279.165	41.011	279.165
2022-01-01	16	0.00	221.4584	37.100	82.296	12.680	89.116	168.56448	0.00	260.7560	66.06976	277.059	55.298	277.059
2022-01-01	17	0.00	157.6448	37.032	78.252	17.376	85.688	133.18080	0.00	258.9600	47.22432	273.467	54.587	273.467
2022-01-01	18	0.00	107.3080	35.876	74.752	27.016	83.300	92.25216	0.00	257.2560	32.11968	271.659	55.236	271.659
2022-01-01	19	0.00	80.4792	35.080	72.480	37.268	83.508	69.18720	0.00	256.5060	24.08960	271.537	57.084	271.537
2022-01-01	20	0.00	64.3864	36.096	75.236	44.544	85.644	55.34976	0.00	257.7050	19.27104	271.715	60.380	271.715
2022-01-01	21	0.00	53.6528	37.028	77.632	49.756	87.520	46.12352	0.00	259.4850	16.06016	271.855	62.984	271.855
2022-01-01	22	0.00	0.0000	49.732	90.688	68.948	98.276	0.00000	0.00	272.8750	0.00000	272.029	76.911	272.029
2022-01-01	23	0.00	0.0000	56.532	92.580	69.140	98.380	0.00000	0.00	274.7220	0.00000	272.231	79.158	272.231
2022-01-02	0	0.00	0.0000	66.812	94.140	76.844	98.760	0.00000	0.04	280.7740	0.00000	272.513	84.139	272.513
2022-01-02	1	0.00	0.0000	71.096	94.212	80.404	98.492	0.00000	0.20	283.7540	0.00000	272.616	86.051	272.616
2022-01-02	2	0.00	0.0000	74.580	94.000	84.192	98.792	0.00000	0.32	287.3460	0.00000	273.024	87.891	273.024
2022-01-02	3	0.00	0.0000	77.660	94.556	86.644	98.980	0.00000	0.48	291.0600	0.00000	273.426	89.460	273.426
2022-01-02	4	0.00	0.0000	93.816	91.588	99.228	100.000	0.00000	0.60	312.8040	0.00000	273.605	96.158	273.605
2022-01-02	5	0.00	0.0000	92.820	88.048	99.388	100.000	0.00000	0.60	311.7720	0.00000	273.537	95.064	273.537
2022-01-02	6	0.00	0.0000	93.304	89.960	99.588	100.000	0.00000	0.68	312.2060	0.00000	273.710	95.713	273.710
2022-01-02	7	0.00	0.0000	93.492	91.856	99.696	100.000	0.00000	0.64	312.9950	0.00000	273.837	96.261	273.837
2022-01-02	8	0.00	0.0000	93.520	93.408	99.756	100.000	0.00000	0.64	313.6920	0.00000	273.885	96.671	273.885
2022-01-02	9	0.85	1.1968	93.872	94.488	99.656	100.000	13.37856	0.64	314.2307	0.53568	274.263	97.004	274.263
⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮
2024-05-19	16	NA	741.68400	3.748	12.400	13.308	26.132	184.14656	0	315.8730	129.28704	300.115	13.897	300.115
2024-05-19	17	NA	653.24720	4.836	13.920	15.876	30.608	177.16928	0	315.2560	119.13024	297.226	16.310	297.226
2024-05-19	18	NA	551.01840	6.664	17.332	22.384	39.596	171.59360	0	315.9100	104.58816	293.967	21.494	293.967
2024-05-19	19	NA	451.50800	7.392	17.692	33.540	48.368	157.55328	0	315.3600	87.64096	290.504	26.748	290.504
2024-05-19	20	NA	365.32160	8.288	16.824	44.928	56.940	132.32256	0	314.6400	70.99328	287.699	31.745	287.699
2024-05-19	21	NA	304.43328	7.720	14.724	52.296	62.416	110.26688	0	313.1420	59.15904	286.748	34.289	286.748
2024-05-19	22	NA	0.00000	0.932	0.144	86.084	86.800	0.00000	0	302.1770	0.00000	286.008	43.490	286.008
2024-05-19	23	NA	0.00000	0.544	0.144	75.572	76.660	0.00000	0	300.4080	0.00000	285.319	38.230	285.319
2024-05-20	0	NA	0.00000	0.364	0.300	64.076	65.364	0.00000	0	298.4437	0.00000	284.668	32.526	284.668
2024-05-20	1	NA	0.00000	0.268	0.300	52.524	53.820	0.00000	0	296.7120	0.00000	284.064	26.728	284.064
2024-05-20	2	NA	0.00000	0.216	0.304	44.528	45.840	0.00000	0	295.4120	0.00000	283.597	22.722	283.597
2024-05-20	3	NA	0.00000	0.180	0.464	39.212	40.740	0.00000	0	294.4750	0.00000	283.175	20.149	283.175
2024-05-20	4	NA	0.00000	0.104	3.632	33.512	37.208	0.00000	0	289.2320	0.00000	282.832	18.614	282.832
2024-05-20	5	NA	0.00000	0.260	3.596	43.000	46.432	0.00000	0	289.8500	0.00000	282.536	23.322	282.536
2024-05-20	6	NA	3.35600	0.904	5.816	39.872	45.740	4.07168	0	290.3950	0.77888	283.660	23.083	283.660
2024-05-20	7	NA	35.98848	1.968	7.488	33.316	41.208	23.18208	0	290.5790	9.36448	288.299	20.995	288.299
2024-05-20	8	NA	88.90048	3.752	11.000	29.284	41.020	48.74688	0	293.7390	20.81152	292.868	21.264	292.868
2024-05-20	9	NA	148.04160	5.272	15.512	26.252	42.420	77.78240	0	297.8070	31.73120	296.414	22.364	296.414
2024-05-20	10	NA	623.06000	16.520	34.476	12.796	45.516	239.66880	0	318.9140	109.52576	300.288	27.327	300.288
2024-05-20	11	NA	697.54480	10.808	32.828	11.096	43.596	243.46960	0	321.0400	118.63680	303.036	24.582	303.036
2024-05-20	12	NA	757.79520	10.524	30.456	11.920	42.148	243.22688	0	322.7000	125.46112	304.884	23.762	304.884
2024-05-20	13	NA	786.57280	12.068	29.524	13.816	41.716	251.42784	0	325.2360	128.43776	305.034	24.281	305.034
2024-05-20	14	NA	780.91120	14.284	31.840	17.044	45.484	270.32896	0	329.1890	127.48992	303.220	27.163	303.220
2024-05-20	15	NA	748.39680	17.028	35.120	22.488	52.052	294.02496	0	333.3480	123.05600	300.109	31.672	300.109
2024-05-20	16	NA	428.61520	34.728	57.824	49.888	92.296	423.56096	0	358.2960	79.20320	297.543	58.684	297.543
2024-05-20	17	NA	367.40000	34.492	62.112	44.068	93.800	396.17088	0	359.0440	69.55136	295.564	58.618	295.564
2024-05-20	18	NA	312.24160	34.684	66.848	42.728	94.424	354.52992	0	358.4140	60.48000	293.613	59.671	293.613
2024-05-20	19	NA	261.56960	31.080	66.276	49.852	93.684	303.41760	0	354.7380	51.42720	291.316	60.223	291.316
2024-05-20	20	NA	212.68736	30.616	64.848	57.588	93.680	249.77856	0	351.5110	41.86048	288.794	61.683	288.794
2024-05-20	21	NA	177.24096	29.016	62.472	63.936	94.160	208.14784	0	347.9070	34.88192	288.011	62.396	288.011

date	hour	production	dswrf_surface	uswrf_top_of_atmosphere	csnow_surface	dlwrf_surface	uswrf_surface	tmp_surface	hourly_cloud_average	hourly_max_t	is.weekend	is.ramadan	is.religousday	is.nationalday	is.publicholiday
<IDate>	<int>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>
2022-01-01	4	0.00	0.0000	0.00000	0.00	227.9990	0.00000	269.220	6.807	269.220	1	0	0	1	1
2022-01-01	5	0.00	0.0000	0.00000	0.00	227.7740	0.00000	269.104	9.254	269.104	1	0	0	1	1
2022-01-01	6	0.00	0.0000	0.00000	0.00	227.7640	0.00000	269.035	10.449	269.035	1	0	0	1	1
2022-01-01	7	0.00	0.0000	0.00000	0.00	228.1960	0.00000	269.001	16.306	269.001	1	0	0	1	1
2022-01-01	8	3.40	0.0000	0.00000	0.00	228.6570	0.00000	269.002	19.933	269.002	1	0	0	1	1
2022-01-01	9	6.80	7.3688	8.96704	0.00	229.4160	2.41280	271.634	23.986	271.634	1	0	0	1	1
2022-01-01	10	9.38	180.1384	135.20448	0.00	237.2910	59.18848	275.786	46.014	275.786	1	0	0	1	1
2022-01-01	11	7.65	254.7928	152.98944	0.00	237.3870	82.53120	278.553	41.338	278.553	1	0	0	1	1
2022-01-01	12	6.80	312.5520	167.55392	0.00	238.3870	99.73056	280.215	39.690	280.215	1	0	0	1	1
2022-01-01	13	5.10	347.4144	180.69312	0.00	240.8870	109.43168	280.609	40.912	280.609	1	0	0	1	1
2022-01-01	14	5.10	364.6544	187.03872	0.00	243.0150	113.51040	280.277	40.812	280.277	1	0	0	1	1
2022-01-01	15	1.70	362.2984	188.88640	0.00	245.2960	112.10304	279.165	41.011	279.165	1	0	0	1	1
2022-01-01	16	0.00	221.4584	168.56448	0.00	260.7560	66.06976	277.059	55.298	277.059	1	0	0	1	1
2022-01-01	17	0.00	157.6448	133.18080	0.00	258.9600	47.22432	273.467	54.587	273.467	1	0	0	1	1
2022-01-01	18	0.00	107.3080	92.25216	0.00	257.2560	32.11968	271.659	55.236	271.659	1	0	0	1	1
2022-01-01	19	0.00	80.4792	69.18720	0.00	256.5060	24.08960	271.537	57.084	271.537	1	0	0	1	1
2022-01-01	20	0.00	64.3864	55.34976	0.00	257.7050	19.27104	271.715	60.380	271.715	1	0	0	1	1
2022-01-01	21	0.00	53.6528	46.12352	0.00	259.4850	16.06016	271.855	62.984	271.855	1	0	0	1	1
2022-01-01	22	0.00	0.0000	0.00000	0.00	272.8750	0.00000	272.029	76.911	272.029	1	0	0	1	1
2022-01-01	23	0.00	0.0000	0.00000	0.00	274.7220	0.00000	272.231	79.158	272.231	1	0	0	1	1
2022-01-02	0	0.00	0.0000	0.00000	0.04	280.7740	0.00000	272.513	84.139	272.513	1	0	0	0	1
2022-01-02	1	0.00	0.0000	0.00000	0.20	283.7540	0.00000	272.616	86.051	272.616	1	0	0	0	1
2022-01-02	2	0.00	0.0000	0.00000	0.32	287.3460	0.00000	273.024	87.891	273.024	1	0	0	0	1
2022-01-02	3	0.00	0.0000	0.00000	0.48	291.0600	0.00000	273.426	89.460	273.426	1	0	0	0	1
2022-01-02	4	0.00	0.0000	0.00000	0.60	312.8040	0.00000	273.605	96.158	273.605	1	0	0	0	1
2022-01-02	5	0.00	0.0000	0.00000	0.60	311.7720	0.00000	273.537	95.064	273.537	1	0	0	0	1
2022-01-02	6	0.00	0.0000	0.00000	0.68	312.2060	0.00000	273.710	95.713	273.710	1	0	0	0	1
2022-01-02	7	0.00	0.0000	0.00000	0.64	312.9950	0.00000	273.837	96.261	273.837	1	0	0	0	1
2022-01-02	8	0.00	0.0000	0.00000	0.64	313.6920	0.00000	273.885	96.671	273.885	1	0	0	0	1
2022-01-02	9	0.85	1.1968	13.37856	0.64	314.2307	0.53568	274.263	97.004	274.263	1	0	0	0	1
⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮
2024-05-14	18	0.14	568.46960	169.69920	0	264.016	110.46400	286.9810	10.242	286.9810	0	0	0	0	0
2024-05-14	19	0.00	469.56800	150.26752	0	261.975	94.29376	283.1640	9.776	283.1640	0	0	0	0	0
2024-05-14	20	0.00	379.66336	124.10624	0	259.407	76.52352	279.9370	8.808	279.9370	0	0	0	0	0
2024-05-14	21	0.00	316.38528	103.42080	0	257.020	63.77152	279.0600	8.659	279.0600	0	0	0	0	0
2024-05-14	22	0.00	0.00000	0.00000	0	243.652	0.00000	278.3870	14.836	278.3870	0	0	0	0	0
2024-05-14	23	0.00	0.00000	0.00000	0	242.829	0.00000	277.7550	15.197	277.7550	0	0	0	0	0
2024-05-15	0	0.00	0.00000	0.00000	0	241.878	0.00000	277.1440	14.826	277.1440	0	0	0	0	0
2024-05-15	1	0.00	0.00000	0.00000	0	241.984	0.00000	276.8100	18.349	276.8100	0	0	0	0	0
2024-05-15	2	0.00	0.00000	0.00000	0	243.318	0.00000	276.6900	24.441	276.6900	0	0	0	0	0
2024-05-15	3	0.00	0.00000	0.00000	0	246.041	0.00000	276.9520	30.521	276.9520	0	0	0	0	0
2024-05-15	4	0.08	0.00000	0.00000	0	265.722	0.00000	276.9600	62.846	276.9600	0	0	0	0	0
2024-05-15	5	0.84	0.00000	0.00000	0	268.502	0.00000	276.9870	64.873	276.9870	0	0	0	0	0
2024-05-15	6	2.94	2.19840	4.11968	0	270.013	0.44928	277.9850	66.249	277.9850	0	0	0	0	0
2024-05-15	7	6.58	23.85280	32.98688	0	271.004	4.95872	281.3370	66.061	281.3370	0	0	0	0	0
2024-05-15	8	9.13	70.55808	64.85632	0	271.370	14.99968	285.4930	65.294	285.4930	0	0	0	0	0
2024-05-15	9	9.82	133.17440	93.23584	0	272.017	27.34272	290.0600	64.018	290.0600	0	0	0	0	0
2024-05-15	10	9.87	671.21840	220.83264	0	271.943	122.53632	295.1730	43.818	295.1730	0	0	0	0	0
2024-05-15	11	9.82	770.01600	205.89824	0	270.559	133.23072	299.4520	33.187	299.4520	0	0	0	0	0
2024-05-15	12	7.92	837.92160	199.86368	0	271.397	139.62432	301.1160	24.561	301.1160	0	0	0	0	0
2024-05-15	13	7.18	877.82480	199.39392	0	273.757	143.27296	301.0770	21.357	301.0770	0	0	0	0	0
2024-05-15	14	6.00	891.83360	201.66912	0	276.612	144.44480	300.1656	20.904	300.1656	0	0	0	0	0
2024-05-15	15	3.11	880.27072	207.31840	0	280.244	143.31904	298.0160	23.678	298.0160	0	0	0	0	0
2024-05-15	16	0.00	645.33440	259.09504	0	305.592	117.19168	295.4920	53.864	295.4920	0	0	0	0	0
2024-05-15	17	0.00	563.07520	245.66080	0	305.315	105.66080	292.5800	54.906	292.5800	0	0	0	0	0
2024-05-15	18	0.00	479.35600	223.52064	0	304.388	92.77632	289.5560	52.225	289.5560	0	0	0	0	0
2024-05-15	19	0.00	394.93824	195.49504	0	302.328	77.94688	286.2330	50.264	286.2330	0	0	0	0	0
2024-05-15	20	0.00	319.39200	161.11552	0	301.056	63.09376	283.7850	49.482	283.7850	0	0	0	0	0
2024-05-15	21	0.00	266.16064	134.26176	0	300.140	52.57856	283.2000	48.427	283.2000	0	0	0	0	0
2024-05-15	22	0.00	0.00000	0.00000	0	290.460	0.00000	282.4660	38.033	282.4660	0	0	0	0	0
2024-05-15	23	0.00	0.00000	0.00000	0	286.686	0.00000	281.7430	34.667	281.7430	0	0	0	0	0